Summary

As the Covid-19 virus began to spread, shuttering businesses and upending economies, the Managing Director of the International Monetary Fund (IMF), Kristalina Georgieva, saw opportunity in the recovery from the unraveling crisis. Georgieva had already identified income and wealth inequality as a central challenge facing the global economy. In her view, the pandemic brought both the risk that poverty and inequality would exponentially worsen and a “once in a lifetime opportunity” to invest in recoveries that transform economies to make them “greener, smarter, and fairer.”

In speeches, interviews, and opinion pieces, Georgieva laid out a strategy for how the IMF can help governments emerge from the pandemic with more inclusive, sustainable, and resilient economies, from modernizing tax systems to make them more progressive to investing more in health, education, and renewable energy, restructuring unsustainable debt, and building stronger social protection systems. This vision—what Georgieva called a “new Bretton Woods moment”—carves out a different approach to the one the Fund took following the 2008 Financial Recession and continues a significant evolution in how it sees its role in promoting global economic stability. Although in laying out this vision, Georgieva did not explicitly reference human rights, including the IMF’s human rights obligations, the Fund’s response to the crisis has monumental significance for hundreds of millions of people’s economic, social and cultural rights.

This report analyzes 39 IMF loan programs approved between March 2020, the official start of the Covid-19 pandemic, and March 2023 under three lending facilities that include conditionalities—the Extended Credit Facility (ECF), Extended Fund Facility (EFF), and Stand-By Arrangement (SBA)—to understand how the IMF is approaching the current crisis in practice and to assess the extent to which it aligns with international human rights standards. The total population in the countries affected by these programs is more than one billion people. The report also includes an in-depth case study of Jordan, where the government has been implementing a series of IMF programs since 2012, based on over 70 interviews and a review of relevant documents and data.

The report makes several key findings. First, it finds that the Fund continues to rely on its traditional approach of expecting governments to reduce debt through fiscal consolidation, which raises significant human rights concerns. While “fiscal consolidation” refers broadly to reducing debt by cutting public spending or increasing revenues, the report finds that programs routinely include conditionalities that impose cuts to public spending and raise the tax burden in ways that disproportionately burden people on low incomes, and they do not adequately explore alternative approaches that are more protective of rights. While many loan programs acknowledge likely harmful social impacts and the importance of “protecting the poor and vulnerable,” they do not generally include an analysis of these impacts, despite guidance in its social spending strategy to do so. Under international human rights law, governments and international financial institutions have an obligation to conduct thorough human rights impact assessments of loan programs prior to the provision of the loan and pursue policies based on the outcome of these assessments that best protect and progressively realize, rather than erode, people’s economic and social rights.

Second, the report finds that a central feature of the Fund’s efforts to address the social impact of its loan programs and build more inclusive economies is the establishment of “social spending floors.” These floors, which are distinct from International Labour Organization (ILO) social protection floors, described in more detail below, typically set a minimum amount a country is expected to spend on specified social programs and “social safety nets,” a term that the IMF and World Bank generally use to refer to tax-financed social assistance to “targeted poor and vulnerable people,” rather than rights-aligned social security programs. These floors serve multiple purposes, including protecting specified social programs from spending cuts as part of fiscal consolidation, as well as “mitigating” the social impact of program adjustments and improving countries’ social protection programs.

The Covid-19 pandemic cast a new light on the importance of social protection, as the reality that half the world’s population—some four billion people—are not covered by any form of social protection contributed to both a humanitarian and economic crisis as the virus shuttered businesses and wreaked havoc on livelihoods. Social protection is key to securing a range of human rights, including the right to social security. The United Nations defines social protection as a set of policies and programs aimed at preventing, protecting, and overcoming situations that adversely affect the well-being of people throughout the course of their lives. It encompasses a wide range of programs that provide support particularly in situations that affect people’s ability to earn an adequate income because of social and attitudinal barriers in conjunction with other conditions such as sickness, disability, older age, unemployment, and childrearing. Social protection systems include a mix of contributory schemes, referred to as social insurance, and tax-financed benefits, referred to as social assistance or sometimes as social safety nets.

While the Fund’s increased attention to social spending more broadly, and social protection in particular, are positive steps toward creating the fiscal space for governments to fulfill human rights, the report finds its approach of heavily relying on social spending floors and poverty-targeted cash transfer programs falls far short of fulfilling rights and can serve to obscure the harmful rights impacts of IMF programs.

The International Covenant on Economic, Social and Cultural Rights, ratified by 171 governments, enshrines all people’s rights to health, education, social security, housing, and an adequate standard of living, which includes food, among other rights. While the full realization of these rights is expected to be realized progressively, the covenant requires governments and international organizations to take immediate steps to facilitate their realization including through legislative and other measures, and to use “the maximum of their available resources,” including to provide and seek “international assistance and co-operation.” Failing to do so is a violation of international human rights law. Many IMF conditionalities, sometimes referred to as “adjustments,” demand changes to structural economic policies in ways that can undermine governments’ ability to realize these rights. In some cases, these adjustments can reshape fundamental aspects of the social contract; that is, in political theory, the theoretical tacit agreement between citizens and their government that underpins its legitimacy.

Existing social contracts have yielded explosive inequality and devastated the environment, so the IMF’s recent shift offers a critical opportunity for building new social contracts rooted in international human rights standards. While social protection is a key pillar of a rights-aligned social contract, it works alongside other pillars, such as quality public services and strong labor laws. However, IMF adjustments can erode these other pillars, such as by placing strict limits on public wages, leading to a retrogression of rights. Social spending floors should not substitute for an approach that pursues policies that avoids retrogression and best fulfills rights, including considering alternatives to austerity. Moreover, this report finds numerous gaps in how social spending floors are defined and designed that limit their effectiveness.

In a similar vein, it finds that the Fund’s focus on poverty-targeted cash transfer programs, often in partnership with the World Bank, is far too narrow to protect people’s right to social security. The Fund’s position is out of step with the growing number of international institutions and scholars that have recognized that universal social protection programs, which provide support to everyone in certain groups, are far better at ensuring that all people can access their rights when compared to means-tested programs, which target people based on socio-economic status, such as their income or assets. The World Bank has formally committed to promoting universal social protection, although in practice it is heavily investing in poverty-targeted programs, including in close collaboration with IMF programs.

Jordan is illustrative of how the IMF’s current approach fails to meet the vision that Georgieva laid out at the beginning of the pandemic. Since 2012, the government has enacted sweeping economic reforms as part of a US$4.4 billion series of IMF programs that imposed austerity-driven measures that increased the cost of living. Recent programs included social spending floors that incorporate a new poverty-targeted cash transfer program, called Takaful, that the government established in 2019 with the support of the World Bank and the United Nations Children’s Fund (UNICEF), in part to offset these reforms (The program has since been renamed the Unified Cash Transfer Program). While the spending floor and Takaful were an improvement over the earlier programs, they are a bandage on a bullet wound. The reforms generated billions in savings and new revenues through measures that increased the cost of living, yet public spending on health, education, and social assistance did not increase as a percentage of the budget. The establishment of Takaful has served as an important lifeline, especially during the pandemic, but it covers only about one in five Jordanian families living under the poverty line, a measure which doesn’t fully capture the millions of people whose rights are affected by these reforms, including the large refugee population living in Jordan excluded from the program. To add salt on the wound, Jordan’s debt-to-GDP ratio is now higher than it was when the IMF approved the first program in this series a decade ago.

Protecting Human Rights Amid Economic Crises

Economic crises—and the responses to them—usually fall hardest on people on low incomes, and the global crisis triggered by the Covid-19 pandemic is no exception. The World Bank estimates that 97 million people had their incomes fall below the poverty line in 2020 alone, and the 2022 Global Report on Food Crises reported that a record 193 million people were experiencing food insecurity, an increase of 58 million people since 2019. At the same time, many governments have accumulated significant debt loads due to increased spending and falling revenues related to the pandemic, even as low- and middle-income countries face steep borrowing costs that impede their ability to recover. How the IMF and governments responded to the crisis can exacerbate the devastating impacts on people’s rights, or, conversely, address long-standing gaps that protect and improve the realization of rights.

Efforts to reduce debt by cutting public spending or raising regressive taxes—that is, austerity—can have well-documented harmful impacts on human rights. Recognizing this, international human rights law sets out strict parameters for when and how states may pursue austerity measures, including that they “demonstrate that all other alternatives have been exhausted, and provide safeguards to protect human rights, particularly the rights of the most vulnerable.” Austerity measures must also be temporary, necessary, proportionate, respectful of minimum core human rights obligations, and non-discriminatory. To ensure these conditions are met, governments and international financial institutions are expected to conduct human rights impact assessments that weigh the potential impacts of policy alternatives. These assessments should serve as the basis for a transparent and credible public dialogue that leads to pursuing the policy option most protective of rights and a plan to effectively safeguard rights when policies that are unavoidable present high risks of retrogression. Guidelines endorsed by the United Nations Human Rights Council advise that austerity is a prima facie violation of human rights unless these conditions are met.

Applied to IMF programs, staff should justify why policies included in specific programs are most likely to lead to the fulfillment of rights, such as by reducing poverty and inequality and enabling adequate spending on sectors essential to the realization of rights without discrimination. They should also undertake detailed assessments on the distributional impacts of program conditionalities and how governments are expected to safeguard rights from any harmful impacts. The IMF’s strategy on social spending adopted in 2019 provides an avenue for conducting such assessments by directing staff to “analyze and, as appropriate, document the social impact of adjustment and measures to protect the vulnerable.” Adjustments are structural economic reforms to which countries need to agree in order to secure an IMF loan. However, as described below, IMF loan programs largely lack such analysis and documentation. The IMF should publish these analyses prior to the approval of programs to facilitate an informed and meaningful dialogue with the public as part of loan negotiations.

Analysis of Loan Programs Approved March 2020 to March 2023

Between March 2020 and March 2023, the IMF approved 34 loans under the Extended Credit Facility (ECF) or Extended Fund Facility (EFF) arrangements and six under the Stand-By Arrangement (SBA). Programs under the ECF and EFF are typically three to five years long, while those under the SBA are generally one to three years long; all three lending facilities include conditionalities (This report’s analysis excludes a $15.6 billion program to Ukraine approved in March 2023 given the extenuating circumstances in the country). These programs vary in many respects, but most share a common theme: they preserve the IMF’s traditional approach while increasing the emphasis on “mitigation.” This approach is best captured by the phrase, a variation of which appears in most programs, “fiscal consolidation while protecting social spending.” All but six programs include a social spending floor.

Doubling Down on Austerity

Social spending floors need to be understood in the context of IMF programs, since one of the central functions they serve is to mitigate impacts of structural adjustments. Of the 39 loan programs reviewed, all but six are explicitly underpinned by “fiscal consolidation,” a term that refers to reducing government debt by reducing spending and/or raising revenues. Although it is possible to achieve fiscal consolidation solely through raising revenues in a progressive manner, in practice most programs reviewed rely in part on decreasing public spending as a percentage of the GDP; that is, austerity. Many common measures to reduce spending can have a directly harmful impact on rights, and many economists, including some within the IMF, argue that cutting spending during a crisis can exacerbate poverty and inequality. Efforts to increase revenue can also have harmful rights impact where they overly rely on regressive taxes. Taxes on goods and services, called “indirect taxes,” tend to be regressive. Those with lower incomes or wealth will often spend a bigger percentage of their income on such taxes than those with higher income or wealth.

Many programs make explicit reference to social spending floors as “mitigating” the impacts of program adjustments, but all but one lack detailed assessments of the impacts of these measures, making it impossible to assess whether they are effectively filling this role. In the Ecuador program, which replaced an earlier program that led to Indigenous-led public protests, IMF staff did conduct an analysis of the program measures’ impact by income decile, but even that analysis was missing key elements and surfaced several problems.

This report tracks three common policy measures included in the initial loan agreements: reductions in public wage bills; reducing or eliminating energy subsidies; and increasing revenues through value-added taxes (VAT). It finds 22 programs include structural benchmarks or general advice to lower the public wage bill, generally through freezing hires and capping or lowering salaries. Such reforms can harm governments’ capacity to fulfill their rights obligations by undermining their ability to deliver quality public services. While some programs explicitly exempt the health and education sectors from cuts or caps, few exempt other sectors essential to rights, such as water and sanitation, environmental agencies, and public transportation. Particularly in the context of high inflation, it can also effectively reduce salaries for public sector employees. Reductions in public sector jobs or salaries also disproportionately hurt women both because they are more likely to fill the gap when public social services are absent and because women tend to be overrepresented in public sector jobs.

Twenty programs remove or reduce consumption-based subsidies on fuel or electricity or develop plans to do so. Fossil fuel subsidies direct enormous amounts of public resources to propping up an industry that drives the climate crisis and harms people’s health and the environment; phasing them out is crucial for avoiding the worst impacts of climate change, reducing the burden on government budgets, combatting inequality, and transitioning to renewable energy sources. However, it is critical that fossil fuel subsidy reform is undertaken in concert with other measures, such as adequate investment in social protection and renewables. The wealthy disproportionately capture the benefits of energy subsidies because they consume more, but without adequate protection measures the impact of removing or reducing consumption-based subsidies is felt most acutely by people on low incomes as it forces them to pay a higher share of their income for transport and goods and services that are tied to energy prices and that they must access to realize their rights.

Twenty-three of the reviewed programs include measures to increase VAT rates, remove exemptions, or otherwise increase VAT revenues. Value-added taxes, which are an indirect tax, tend to be regressive since they are the same for people regardless of income. For example, the Sri Lanka program increased VATs from 8 to 15 percent and the Zambia program removed VAT exemptions on most unprocessed foods. Some IMF and other publications have argued that VATs are not regressive because the wealthy pay a greater share of the total revenue collected because they consume more, but the tax consumes a higher share of low-income families’ income, and in some cases, it can make food and other goods essential to rights no longer accessible to them. A study of VATs in France found that the poorest households pay more than 20 percent of their income in consumption taxes, compared to less than 10 percent for the wealthiest households. In many countries, VAT is applied to menstrual health products, which can impede girls from attending school, among other rights harms.

This report’s analysis is largely limited to IMF documents and, with the exception of Jordan, it does not investigate the on-the-ground impact of the programs reviewed; in some cases, governments may not have implemented conditionalities as they appear in program documents. This report does not track other policy measures that can harm the rights of people on low incomes and that can be found in some programs, such as pension reform, labor flexibilization, and exchange rate reforms. At the same time, some include positive reforms for the realization of rights, such as those related to anti-corruption and progressive revenue generation, that are also beyond the scope of this report.

The False Promise of IMF Social Spending Floors

In the context of IMF programs, social spending floors fill multiple functions. They are variously described as intended to protect social sectors from cuts, mitigate impacts of structural adjustments, and develop countries’ social protection programs. Yet nearly all loan programs lack adequate information to properly assess the practical implications of these floors, which undermines the public’s access to information crucial to a meaningful debate about these measures and its ability to meaningfully engage with their government and the IMF to ensure that loan programs protect their rights. For example, most programs do not include baseline information on social spending levels prior to IMF loan programs, making it difficult if not impossible to assess whether and to what extent the floors will lead to an overall increase in social spending. The way in which social spending floors are defined also often makes it difficult to assess which social programs, if any, will benefit from an increase in spending. While adequate spending is not in itself sufficient to guarantee the fulfillment of rights, it is nonetheless necessary.

In many cases the floors are defined to cover broad categories of spending that go far beyond social protection, and encompass spending on health, education, and other social sectors. This is important for protecting as many social programs as possible from potential spending cuts, but it does not alone ensure that people’s rights are safeguarded from the impacts of adjustment, nor does it guarantee investment in improving social protection. For example, Benin’s program defines the floor as covering 72 enumerated programs across 12 sectors, including “integrated management of border areas” and “development of rural roads.” While defining the floor so broadly serves an important purpose of protecting numerous areas of social spending, even if there is an increase in total spending on these areas, it does not necessarily compensate for the impact of adjustments in the program. In Benin, for example, these include removing VAT exemptions for imported rice, water, and electricity, even as one in three children under five suffer from chronic malnutrition, according to the World Food Programme.

To address this problem, the IMF should adopt separate floors to serve distinct purposes: one for protecting and increasing spending in areas critical for fulfilling social and economic rights without discrimination, and another for mitigating against necessary measures that nonetheless pose risk to rights, such as phasing out fossil subsidies. The first should be broadly defined, and disaggregated by sector, and ensure that spending on, for example, health, education, and social protection are, at a minimum, in line with international benchmarks as a percentage of GDP and of national budgets. The second should be based on published assessments of program impacts and demonstrate how the target, such as increased social protection spending or coverage, adequately mitigates those impacts to avoid, at a minimum, the retrogression of rights. For both, the IMF should consider replacing “floors,” which are revised in each review, with “goals” to be achieved by the end of the program, and a plan developed from the outset to achieve specific benchmarks in each review.

Moreover, in most cases, it is not possible to assess whether, or to what extent, the floor increases spending relative to previous years. Where floor definitions track budget lines, it may be possible to independently calculate this, but many floors include categories that aren’t tracked in public budgets. For example, Jordan’s floor covers “non-wage health and education current expenditures,” a category not specifically tracked in the budget, making it difficult to compare the amount provided for in the floor to spending prior to the program. All social spending floors should include baseline information to assess any increase in spending reflected in social spending floors.

Despite the challenges of assessing social spending floors, available information indicates three key problems. First, the IMF’s preference for poverty-targeted—also known as means-tested—cash transfer programs likely leads to setting floors too low to protect rights. Second, inadequate attention is paid to sequencing increases in social protection coverage to take place prior to the impact of adjustments. And finally, almost all social spending floors are incorporated as indicative targets, rather than “performance criteria,” the IMF term for requirements that require a Board waiver if not met, enabling the IMF to release additional funds even if government spending does not meet the floors. An analysis by Oxfam of 17 IMF loans approved in 2020 and 2021 found that around one-third of social spending floors were not met.

The Fund is increasingly turning to poverty-targeted cash transfer programs, frequently incorporated into social spending floors, including to address the social impact of its programs. Of the 39 programs reviewed, 26 referenced social protection, usually as structural benchmarks or as part of social spending floors; in nearly all cases, measures and advice were explicitly for targeted programs. These programs frequently replace other types of public programs meant to provide people with support, essentially reshaping existing social contracts. In some cases, improvements to tax-financed social protection schemes came alongside measures that could erode contributory schemes. For example, in Jordan, where the social protection floor incorporates spending on a new poverty targeted cash transfer program, the program also includes measures to exempt certain employers to reduce social security payments for certain employees. In other cases, such as subsidies on fuel or electricity, removing or reducing existing programs is necessary to shift toward a more inclusive and sustainable social contract, but it is critical that these subsidies be replaced by new programs that advance rights and enable a rapid transition to renewable sources of power transportation and electricity for all. However, the IMF’s persistent preference for targeted programs powerfully shapes and limits the role of social protection in evolving social contracts.

Research has shown that poverty-targeted programs have numerous flaws that can seriously hamper governments’ ability to guarantee rights, including from retrogression due to IMF-driven reforms. These include high rates of inaccuracy both due to including those who are not eligible and excluding those who are; bureaucratic and other hurdles to access; the expense of targeting programs; breeding social resentment that undermines support; and being more prone to corruption. Moreover, while these programs often set a goal of covering those living under the poverty line, they often cover only a portion of that group, such as in Jordan and Egypt. But even where they do cover all households under the poverty line, as IMF advice often emphasizes, it is still problematic because these lines are often set so low that they exclude many who are unable to realize their rights, particularly in the context of economic crisis and IMF-driven adjustments. Recognizing some of these problems, recent IMF guidance on social safety nets advises that in countries with limited data, rather than focusing on targeting the benefits, staff should look to “enhance tax capacity and effort” to “claw back” some of the benefits that accrue to high-income households.

In 2008, the body of human rights experts charged with interpreting the ICESCR issued a General Comment noting that “all persons should be covered by the social security system,” which it states necessitates the establishment of non-contributory schemes “in order to achieve universal coverage.” ILO Social Protection Floors Recommendation No. 202, approved by all member countries in 2012, calls on member states to establish and fund social protection floors that extend social security to all. Based on the principle of universality, the floor guarantees access to health care and provides basic income security for children; those unable to work due to social and attitudinal barriers in conjunction with sickness, disability, older age, or unemployment. As noted, the World Bank formally embraced universal social protection in 2016, although in practice it continues to support targeted programs.

The IMF’s program in Ecuador offers a clear example of the problem with the IMF’s preference for targeting safety nets only to those living under the poverty line, as well as inadequate attention to sequencing and enforcement. In addition to including an analysis of the impact of reforms by income decile, Ecuador’s program is unusual in that its floor is defined not as a minimum amount of spending but as a coverage target: the expansion of cash transfers to at least 80 percent of the poorest 30 percent of the population. These unusual features are likely due to public protests in 2019, led by Indigenous groups, opposing a previous IMF program that had sought to remove fuel subsidies without appropriate measures to mitigate its impacts on people on low incomes, among other measures. The subsequent program approved in 2020 included similar measures but with an explicit effort to assess and compensate for costs by expanding the social safety net as part of a World Bank-funded project. Ultimately, some of the measures anticipated by analysis IMF staff conducted were not fully implemented, but the program nonetheless offers a rare opportunity to examine how Fund staff approach social impacts.

The IMF’s analysis calculated that Ecuadorians who did not receive additional cash assistance would experience an effective net decrease in their income due to reforms mandated in the program ranging from 2.1 to 6.3 percent. The target of expanding the social safety net to the poorest 30 percent of the population was set to roughly align with the percentage of the population living under the poverty line prior to the pandemic. However, the pandemic temporarily pushed around 1.5 million people into poverty, increasing the number of people living under the poverty line to 38 percent of the population instead of 30 percent. This means that even if coverage had been immediately expanded in line with IMF expectations, it would have left out roughly 2.5 million people experiencing poverty at a time of economic crisis and rising prices. But coverage was not expanded immediately, and in fact took longer than expected to reach the targets the IMF program set. The government faced challenges in reaching particularly the poorest Ecuadorians; an IMF program review found that by October 2021 it had achieved only 30 percent coverage of the poorest 10 percent. In June 2022, Indigenous-led protests once again broke out to demand economic reforms, including in response to fuel subsidy removal imposed by the IMF. The protests, which forced the government to temporarily relocate to a city outside the capital, ultimately led the government to restore some subsidies.

IMF in Jordan: “I’m Dying of Hunger and I’m Not Getting Aid”

The IMF’s approach of using social spending floors in combination with targeted cash transfer programs to mitigate the impact of program adjustments did not begin with the Covid-19 pandemic. It has been a hallmark of several major programs over the past decade, including the latter part of a decade-long IMF-driven reform of Jordan’s economy. The first program, approved in 2012, removed fuel subsidies with little support to offset the increase in prices, sparking nationwide protests. According to the IMF, subsidy removal led to 5 percent inflation, yet the program did not include a social spending floor. The second program led the government in 2018 to remove bread subsidies and implement tax reforms that raised income taxes on the middle class and increased sales taxes, measures that led to inflation peaking at 5.7 percent in July 2018 and also sparked protests that culminated in the ouster of the prime minister. That program did have a social spending floor as an indicative target, but the government did not meet it in any of the program’s reviews. It provided only limited one-time subsidies to supplement Jordan’s barebones social protection system that at the time provided small amounts to certain groups of people such as orphans and older people. A government report acknowledged that investment in social assistance decreased between 2011 and 2017 despite the subsidy removal generating 788 million JD (US$1.1 billion) in savings during that time. A change in how Jordan calculates poverty in its Household Income and Expenditure Survey between 2011 and 2018 means it is not possible to compare poverty rates over this period, but an independent analysis found an increase of 8 percentage points. Following the 2018 reforms, World Bank data indicates a steady increase in food insecurity, from 12.7 percent in 2017 to 17 percent in 2020, the last year for which there is data.

In 2019, the government, working with the IMF, World Bank, and UNICEF, established a targeted cash transfer program called Takaful that was meant to protect people from the costs of these adjustments. Takaful exemplifies both the power of the IMF’s attention to social protection and the limits of its preference for targeted programs. The timing of the program’s implementation was fortuitous; the first disbursement went out just as the pandemic hit. Originally intended to reach 85,000 households, authorities were able to quickly supplement the program to reach 240,000 informal workers in 2020 (“Takaful-2”) and a total of 160,000 in 2021 (“Takaful-3”). Human Rights Watch interviewed several people who benefitted from Takaful who described how critical the assistance was for being able to pay for food and other necessities.

At the same time, the program was not sufficient to ameliorate price increases that began in 2012 and is far too narrow to ensure people’s rights. People described the challenges they had absorbing the price increases to which IMF-driven reforms contributed with little or no support. The Covid-19 expansion expired at the end of 2021, and the original Takaful was expanded to reach 120,000 beneficiary households in 2022, that amounts to around 5 percent of Jordan’s population of roughly 11 million. Since the start of the pandemic, poverty has increased from 15 to 24 percent, and even at its larger size, Takaful only reached around one in five Jordanians living under the poverty line.

To select beneficiaries, the government first vets whether applicants meet basic eligibility criteria, such as whether the head of the household is a Jordanian citizen with a national ID number. For households that make the cut, the government applies an algorithm to estimate their income based on 57 socio-economic indicators (such as household size, asset ownership, the gender of the head of household, and dwelling characteristics), ranking them from least poor to poorest. The National Aid Fund (NAF) targets cash transfers to those ranked the poorest until all program slots are taken.

Separate Human Rights Watch research published in June 2023 found that some of the indicators resulted in hairsplitting distinctions to determine who should receive benefits or reinforced discriminatory Jordanian policies. Human Rights Watch also found that the algorithm relies on inaccurate data about people’s informal sources of income and living expenses to estimate people’s income. A survey conducted by UNICEF found only marginal differences for rates of food insecurity and other measures among Takaful beneficiaries, those who are eligible but were not selected as beneficiaries, and those who applied but were deemed ineligible. Moreover, non-Jordanian households are not eligible for Takaful despite experiencing higher levels of poverty than Jordanians and families of Jordanian women married to non-Jordanian men are less likely to qualify. As a universal right, the right to social security applies to all, regardless of citizenship or immigration status, in line with the rights to equality and nondiscrimination. 

Human Rights Watch spoke to several people who were rejected by Takaful despite having trouble feeding their family. Rahma, an agricultural worker with six children, told Human Rights Watch she applied twice for assistance and was rejected both times. She and her husband each earn 10 JD (US$14) for a day’s work; the minimum wage in Jordan is 260 JD ($367) per month and a living wage for a typical family, according to one calculation from 2020, is around 600 JD ($846) per month. “We’re basically starving. There was a period of time when we only had two bags of rice and we’d ration it.” People also described being increasingly unable to pay for their electricity bills. “Bassam” has seven children and his wife doesn’t work, but he was denied support because he has a restaurant registered in his name, although it has since closed down. He described his difficulty paying his electricity bills: “They cut off electricity, we go beg for money, and get it turned back on. After three months of not paying, it’s cut off. We sit in candlelight and don’t use food that needs refrigeration.” The most recent IMF program, approved in 2020, led to an overhaul of the electricity pricing system aimed at reducing costs for businesses, which had been paying higher prices to cross-subsidize household tariffs. The reforms incorporate subsidies that, at least for now, appear to protect most Jordanians from tariff increases, but some may still see their bills go up and Jordan’s large refugee population is not eligible for the subsidy.

Jordan’s experience of Takaful carries important lessons for the IMF as it replicates the approach it took there in other countries. A social spending floor, even when combined with a targeted cash transfer program, is not nearly enough to protect people from the impacts of austerity measures or to differentiate the IMF’s Covid-19 response from its previous response to crises.

 

Glossary

Adjustments: Changes to economic policies to which countries agree as a condition for IMF lending.

Austerity: A set of reforms within the context of a contraction of public spending in terms of GDP or in real terms.

Conditionality: Specific terms or actions which governments are required to undertake or achieve in order to receive IMF funds.

Extended Credit Facility: 3- to 5-year IMF loan program available to low-income countries, currently at zero interest.

Extended Fund Facility: 3- to 4-year IMF loan program, subject to interest and other fees.

Fiscal consolidation: Cutting public spending or increasing revenues to reduce deficits or debt.

Indicative target: Flexible quantitative indicators to monitor progress in IMF loans; does not require a waiver by the IMF Board if unmet.

Performance criteria: Specific, measurable conditions in IMF loans that borrowing governments are required to meet or obtain a waiver from the IMF Board.

Prior actions: Steps a government agrees to take before the IMF approves financing or completes a review.

Public wage bill: Total compensation to government employees, including salaries, allowances, and social security contributions.

Social assistance: A subset of social protection comprising tax-financed transfers (that is, not including programs funded through individual or employer contributions).

Social protection: A set of policies and programs aimed at preventing, protecting, and overcoming situations that adversely affect the well-being of people throughout the course of their lives.

• Poverty targeted/means-tested: Eligibility for benefits is based on falling below a certain threshold of income or assets.
• Universal: Benefits are available to everyone within specific groups that encompass stages of life or statuses in which people’s economic, social and cultural rights are particularly at risk, such as children, people with disabilities, unemployed adults, caretakers, and older people.

Social safety net: The term has no clear set definition. The ILO uses the term to describe programs meant to temporarily buffer people from economic shocks, while the World Bank typically uses it to refer to tax-financed social assistance.

Social security: Enshrined in international human rights law, a range of programs, whether funded from contributions or through general taxation, that encompass at least nine areas of support: health care, sickness, older age, unemployment, employment injury, family and child support, maternity, disability, and survivors and orphans.

Social protection floor: Nationally defined sets of basic social security guarantees codified in ILO Recommendation No. 202.

Social spending floor: In IMF programs, minimum amount of spending on specific social programs or ministries or other related measures, such as minimum number of households covered by social assistance programs.

Stand-By Arrangement: Short-term IMF loan program, typically 12-24 months but no longer than 36 months, most used by advanced and emerging market countries.

Structural benchmarks: Reform measures included in IMF programs that are not quantifiable such as enacting specific legislation or undertaking audits.

Value-added taxes: A consumption tax on goods and services that is levied at each stage of the supply chain where value is added

Recommendations

This report demonstrates the inadequacy of the IMF’s current approach to social spending floors both for addressing longstanding gaps and for offsetting the harmful impact on human rights of IMF-driven reforms. While recommendations for specific economic or fiscal reforms are beyond the scope of this report, there are a number of key actions that are critical for pursuing recoveries that advance human rights in the short and long term.

For the IMF

• Prioritize progressive revenue generation and other measures over cuts to public spending when seeking to reduce fiscal deficits and debt.
• Conduct human rights impact assessments of programs that:
  • Demonstrate efforts to avoid austerity measures and establish that any such measures are temporary, necessary, proportionate, respectful of minimum core human rights obligations, and non-discriminatory.
  • Calculate the expected and actual impact of program conditionalities by income decile, as well as on groups that face discrimination, in line with the 2019 social spending strategy. To ensure effective assessments, the IMF should:
  • Explicitly prohibit discrimination in all its operations and seek to identify, end, and remedy any direct or indirect discriminatory impacts of its policies, including based on gender, nationality, age, disability, race or ethnicity, religion, or other grounds prohibited in international human rights law.
  • Periodically review and publish assessments throughout the duration of loan programs.
• Facilitate informed and meaningful engagement with the public in countries with ongoing loan negotiations, particularly with those most likely to be harmed by program conditionalities, including by publishing impact assessments well in advance of program approval.
• Any reforms to public wage bills should ensure adequate staffing at living wages for sectors key to the fulfillment of human rights, including health and education, and consider existing national staffing shortages against international benchmarks.
• Ensure any fossil fuel subsidy reform is structured to advance the realization of people’s rights and the transition to renewable energy. This should include measures, such as expanding social protection, that are implemented prior to subsidy removal, and increased investment in renewable energy that reduces costs, particularly for people on low incomes.
• Consistently include two separate floors in loan programs: One should protect and increase spending in areas critical for fulfilling economic and social rights without discrimination and should be broadly defined, disaggregated by sector, and ensure that spending on, for example, health, education, and social protection are, at a minimum, in line with international benchmarks as a percentage of GDP and of national budgets. A second should adequately mitigate against necessary measures that nonetheless pose risk to rights, such as phasing out fossil fuel subsidies, based on published assessments of program impacts, and should demonstrate how the included measure, such as increased social protection spending or coverage, avoids, at a minimum, the retrogression of rights.
  • Consider replacing “floors,” which are revised ad hoc in each review, with “goals” to be achieved by the end of the program, and a plan developed from the outset to achieve specific benchmarks in each review.
  • Include baseline data for social protection and spending floors, as defined in individual programs.
  • All minimum spending amounts should be included as percentage of GDP as well as in nominal terms.
• Ensure the floor specifically for social protection coverage is set high enough and appropriately defined to adequately offset adjustments and enable broad social protection programs. Specifically, develop standardized design procedures so that floors cover all elements of the ILO social protection floors recommendation and Convention 102 on minimum standards of social security.
• Formally adopt a goal of achieving universal social protection in line with the World Bank and ILO positions.
• Ensure that any reform to social security is in accordance with international labor standards and agreed by all relevant social groups through national social dialogue.
• Protect and strengthen other areas important for the fulfillment of economic and social rights, including public services; labor rights, such as freedom of association and living wage; and governments’ capacity to tackle corruption and tax evasion.
• Commit formally to recognize a duty to respect, protect and fulfil all human rights, including economic and social rights, in all its work, without discrimination.

For IMF Executive Directors

• Support guidance to expect IMF staff to conduct distributional impact analyses for all programs and include them in staff reports.
• Support a careful assessment of the impacts of the IMF’s current approach to lending and possible alternatives and publish the results of the assessment that supports an approach most effective at reducing poverty and inequality and fulfilling economic and social rights in the short- and long-term.

World Bank

• Undertake a review of projects on safety nets to assess their consistency with the World Bank’s stated position on universal social protection and with human rights standards.
• Write and approve a strategy to accelerate action to achieve universal social protection.

For Jordan and other Governments

While the report focuses on Jordan as a case study, these recommendations may apply to other governments pursuing similar economic and fiscal programs. In particular:

Social Spending and Cash Transfers

• Publicly commit to substantially scaling up social spending including on social protection to ensure the protection of the right to social security and other human rights for everyone in the country and ensure that such spending is not decreased as part of fiscal consolidation.
• Transition to universal social protection by phasing in specific groups of people to receive universal benefits, such as children, older people, people who are unemployed, and people with disabilities. During the transition period, remove discriminatory barriers to Takaful by including all residents regardless of nationality, including families of people with temporary Jordanian passports, non-citizen spouses, and children of Jordanian women.
• Ensure dedicated budget lines for both Takaful’s operations and cash transfers in the government budget.

Electricity Reform

• Publicly recognize the right of everyone in the country to affordable and accessible clean electricity.
• Conduct review of economic accessibility of consumer electricity prices and the assumption that usage accurately correlates with income. Ensure electricity remains accessible for low-income households.
• Substantially increase investment in renewable energy for low-income households and other initiatives to increase efficiency and reduce electricity costs central to any future electricity pricing reform.
• Ensure that non-Jordanian citizens, including refugees and migrant workers, have access to affordable electricity.

Progressive revenue generation

• Scale up efforts to halt corruption including by removing any barriers to investigation or prosecution.
• Invest adequate resources to investigate cases of tax evasion on the highest earners.
• Increase revenues through progressive income tax measures that ensure those with the highest income and/or wealth pay their fair share.

Labor rights

• At a minimum, refrain from weakening labor rights such as by reducing employer social security contributions for certain employees.
• Establish an hourly living wage that also extends to informal workers.
• Allow the creation of independent, elected trade unions in all trades, including domestic workers and for migrant workers, limit official involvement in the creation and management of such unions, and allow for the reinstatement of the unions that have been arbitrarily dissolved or dissolved based on abusive or overbroad provisions. Ensure full respect of freedom of association for all trade unions.

 

Methodology

This report analyzes the International Monetary Fund’s use of social spending floors in its loans in the wake of the Covid-19 pandemic and seeks to assess their adequacy for mitigating the impact of program reforms on people on low incomes. It situates this analysis against the backdrop of a significant evolution in the IMF’s rhetoric around poverty and inequality starting in the 1990s up to 2023, as well as its policies and guidance, that articulate an institutional goal of building more inclusive economies and reducing inequality.

The report provides an analysis of key measures, with a focus on social spending floors, in 39 loans approved between March 2020 and March 2023 under the IMF’s Extended Credit Facility, Extended Fund Facility, and Stand-By Arrangement, all of which are lending instruments that include conditionalities. This includes programs in the following countries: Afghanistan (frozen), Argentina, Armenia, Bangladesh, Barbados, Benin, Cabo Verde, Cameroon, Chad, Congo, Costa Rica, Democratic Republic of Congo, Ecuador, Egypt, Gabon, Gambia, Georgia, Guinea-Bissau, Jordan, Kenya, Madagascar, Malawi, Mauritania, Moldova, Mozambique, Nepal, Niger, Papua New Guinea, Senegal, Serbia, Seychelles, Somalia, Sri Lanka, Suriname, Tanzania, Uganda, Ukraine, and Zambia (a loan program to Ukraine approved in March 2023 is not included in this report’s analysis given the extenuating circumstances in the country at the time of approval). The analysis examines the following elements of loan agreements: fiscal consolidation; reforms to public wage bills; subsidy reforms; reforms related to value-added taxes; and references to social spending or social protection and related conditionalities.

It also provides an in-depth analysis of four IMF loan programs in Jordan starting in 2012, where social spending floors, and a related cash transfer program, were intended to offset the impact of IMF-driven fiscal policy and subsidy reforms.

In researching this report, Human Rights Watch reviewed thousands of pages of IMF loan documents, as well as numerous policies, guidance notes, research papers, and institutional reviews. For the case study on Jordan, Human Rights Watch conducted most of the interviews in person in October 2021 with 30 Jordanians (20 men and 10 women) on low incomes living in Amman, Jordan’s capital; Zarqa, a town around 30 kilometers northeast of Amman; and the southern al-Ghor, an agricultural area near the Dead Sea. Human Rights Watch conducted additional interviews with people on low incomes between November 2022 and January 2023.

In addition, Human Rights Watch interviewed representatives from Jordanian and international civil society and trade unions, IMF staff, former Jordanian government officials, relevant UN agencies, humanitarian workers in Jordan, and experts on Jordan’s economy and electricity sector. Human Rights Watch wrote to the IMF on August 9, 2023, the World Bank on July 27, 2023, and the Jordanian and Ecuadorian governments on June 18, 2023, and June 9, 2023, respectively.

In a response to Human Rights Watch’s letter, the World Bank reaffirmed its commitment “to advocate for and support the goal of universal social protection,” but noted that it supports means-testing as part of its strategy to advance toward that goal: “In budget constrained contexts, our experience shows that targeted interventions can be an effective method to distribute a limited amount of resources to those who need it most.” It also notes that while “ the World Bank does work closely with all international partners to share global understanding and best practices on the design and implementation of social protection programs, policies, and strategies,” it “does not have a formalized arrangements in the design of [IMF] programs.” Finally, it lauds Jordan’s cash transfer program as “the largest such program in the Middle East and North Africa Region (MENA) in terms of coverage of the poorest people,” and says that “the government covers this funding using the same mechanisms adopted to finance various other programs in the general budget.

On September 22, in response to Human Rights Watch’s letter, the IMF wrote: “Prioritization of social spending with a focus on supporting the most vulnerable is an important aspect of the IMF’s work and a guiding principle of our 2019 Strategy for Engagement on Social Spending.”

Human Rights Watch informed all interviewees of the purpose of the interview, its voluntary nature, and the ways in which the information would be collected and used. Human Rights Watch assured participants that they could end the interview at any time or decline to answer any questions, without any negative consequences. All interviewees provided informed consent to participate in the interviews. Human Rights Watch did not offer any remuneration. Most names have been changed to protect interviewees.

 

I. Background: IMF and Social Protection

As the Covid-19 virus began to spread, shuttering businesses and upending economies, the Managing Director of the International Monetary Fund, Kristalina Georgieva, saw opportunity in the recovery from the unraveling crisis. Georgieva had already identified income and wealth inequality as a central challenge facing the global economy.[1] In her view, the pandemic brought both the risk that poverty and inequality would exponentially worsen and a “once in a lifetime opportunity” to invest in recoveries that transform economies to make them “greener, smarter, and fairer.”[2]

In speeches, interviews, and opinion pieces, Georgieva laid out a strategy for how the IMF can help governments emerge from the pandemic with more inclusive and resilient economies, from modernizing tax systems to make them more progressive to investing more in health, education, restructuring unsustainable debt, and building stronger social protection systems.[3] This vision—what Georgieva called a “new Bretton Woods moment” —continues a significant evolution in how the Fund sees its role in promoting global economic stability.[4]

The IMF’s Evolving Position on Social Issues

The IMF was established in 1944 as part of the Bretton Woods Agreement, which sought to rebuild the international economic system ravaged by World War II. Its stated purpose, set out in its Articles of Agreement, is to promote international monetary cooperation and bolster the economic stability of member countries.[5] It does this primarily by making concessional loans available to governments facing balance of payment problems (that is, when public revenues fall too far below expenditures) and by issuing regular surveillance reports that assess the financial health of its members, called Article IV reports after the section in the Articles of Agreement that mandates them.

Traditionally, the Fund took a narrow view of its mandate, often invoking a clause in its Articles of Agreement stipulating that Fund guidance with respect to exchange rate policies “shall respect the domestic and social policies of members” to prevent the Fund from engaging on issues such as adequacy of social spending, corruption, and economic inequality.[6] However, this view began to gradually shift in the 1990s, a period when the Fund was heavily involved in the transition of post-Soviet states to market economies.[7] The Fund’s role in promoting “shock therapy,” particularly in Russia, was heavily criticized for removing price controls and implementing other measures in a way that has been blamed for increasing the immiseration of the population while rapid privatization through opaque and corrupt deals created a handful of oligarchs, such that by 1993 the United States formally ended its support for the approach.[8] In the mid-1990s, the managing director issued a series of memos to staff calling for broader IMF engagement in social and distributional policies.[9] At the same time, the Board adopted a number of guidelines for staff engagement on issues considered outside its traditional focus, including on poverty (1991); implications of policies on income distribution (1996); and governance (1997).[10] In 1996, the Board also approved guidelines for collaborating with the ILO on labor market and social protection issues.[11] And beginning in 1998, Article IV surveillance reports were expected to include tables with basic social data.[12]

These developments increased IMF attention to social issues in surveillance reports, the annual reports that monitor and assess a member country’s economy. However, they had limited impact on operations. A 2004 internal review of a range of IMF reports produced between 2002 and 2004 found that while half of Article IV reports included “some discussion of social issues,” references were often superficial and substantive discussions were mostly limited to “post-conflict countries and countries with endemic political and social instability.”[13] The review, referencing the Articles of Agreement clause on respecting the domestic and social policies of members, recommended staff be more selective in when they engage on social issues, to cover “fewer issues in greater depth, perhaps every few years.”[14]

Similarly, the review found that while governance was addressed in 70 percent of surveillance reports, they “tend to be limited to descriptions of facts and of the authorities’ past or current actions; they tend to stop short of assessing adequacy of these steps or needs for further action.”[15] Moreover, corruption—the underlying problem governance is meant to address—was only referenced in 10 percent of reports.[16] In other words, while staff were increasingly paying attention to issues like social protection and governance in their reports, they rarely made their way into policy prescriptions embedded in IMF programs. In response to the review, the IMF’s Board of Directors were “broadly satisfied” with how governance issues were being monitored, although some expressed support for more research and analysis on how to improve growth and resilience for low-income countries and more closely monitoring progress toward the Millennium Development Goals.[17]

The aftermath of the 2008 recession laid bare the limits of the Fund’s approach to social issues. Although during the crisis the IMF supported increased spending, including on social sectors, by 2010 that gave way to insisting that governments seeking bailouts drastically contain their debt by reducing their spending, including on programs vital to protecting people’s rights.[18] A comprehensive review of IMF programs by the ILO and Columbia University in 187 countries between 2010 and 2020 demonstrated how austerity-driven programs gutted funding for health, education, pensions, and social assistance programs, at a time when huge sectors of the population were reeling from job and income losses due to the recession.[19]

In Europe, the push for austerity, particularly in Greece, Spain, Ireland, and Portugal, where the EU also drove austerity policies, faced massive popular backlash, badly damaging the Fund’s reputation.[20] This fallout worsened when the bitter medicine the Fund had prescribed was not working to heal ailing economies. On the contrary, the economic results were often disastrous, in some cases leaving countries mired in deepening poverty and inequality.[21] In January 2013, the IMF’s chief economist, Olivier Blanchard, issued a paper that amounted to a rare acknowledgment that the Fund’s insistence on austerity was deeply flawed. The paper compared economic growth in countries where austerity was imposed to IMF projections and found that austerity had the exact opposite effect than the IMF had anticipated.[22] Remy Davison, a professor on international relations at Monash University, summed up the paper’s findings: For every dollar governments cut from public spending, the economy shrank by $1.50, whereas the Fund had predicted that for every dollar cut in government spending, economic output would only shrink by 50 cents.[23] The paper said the IMF underestimated the “fiscal multiplier effect;” the extent to which government spending effects levels of consumer spending and investment.

The IMF’s response to the recession set off a vigorous debate about its economic approach, and renewed calls, including by many IMF staff, for the need to take the social and political implications of its programs much more seriously. This was, in part, an exercise in repairing its reputation. Christine Lagarde, who was appointed managing director in 2011, made it a point to, in her words, give the IMF a “human face,” and spoke more often about issues like gender and income inequality.[24] In 2012, she gave a speech specifying “better social protection” as one way of achieving more “inclusive economies.”[25]

This shift also made its way into IMF research, guidance, and reporting. In 2012, the IMF’s internal guidance changed its definition of “macro-criticality,” meaning that it affects a country’s macroeconomic situation, which is the standard for staff to assess whether issues are within its mandate. The new definition includes any issue that “affects, or has the potential to affect, domestic (e.g., growth and inflation) or external stability.”[26]Staff maintained broad discretion over the issues they chose to work on, but this gave them a stronger basis to tackle emerging issues outside their traditional focus.

The IMF Board approved a crop of new policies and guidance notes that focused on these emerging issues. For example, in 2013, the Fund issued new guidance on jobs and growth that called for more systemic integration of policy advice to encourage, among other things, reduced inequality and better protection of the “vulnerable.”[27] A review of 2002 conditionalities the following year revised existing guidance to instruct staff that, “if feasible and appropriate, any adverse effects of program measures on the most vulnerable should be mitigated.”[28] Previously, even after establishing that there was a “strong justification” for working on issues outside the IMF’s core mandate, staff were expected to rely on the expertise of institutions such as the World Bank and had little recourse where such expertise was lacking or inadequate.[29]

In 2018, the Board approved new guidelines on social safeguards and a guidance note for improving the Fund’s engagement on governance, both of which gave staff more leeway to address social issues when designing programs.[30] The same year, IMF staff published two reports on operationalizing inequality and gender in country work.[31] In 2019, the Board approved a new strategy on social spending, which is discussed in detail in the following section.[32]

The IMF and Social Protection

Social protection is broadly recognized, including by the IMF, as key for reducing poverty and inequality.[33] This is closely linked with the human right to social security, which first appeared in the Universal Declaration of Human Rights and was later enshrined in the International Covenant on Economic, Social and Cultural Rights, ratified by 171 countries.[34] The Committee charged with monitoring the treaty’s implementation has interpreted this right to encompass nine areas of support: health care; sickness; older age; unemployment; employment injury; family and child support; maternity; disability; and survivors and orphans following the death of a breadwinner.[35]

Social protection is a less well-defined concept. The United Nations defines it as a set of policies and programs aimed at preventing, protecting, and overcoming situations that adversely affect the well-being of people, throughout their life course.[36] Most often this includes a combination of cash or in-kind (such as food) transfers distributed through contributory or non-contributory schemes.[37] This report uses the UN definition, but the term is sometimes used to encompass a broader range of public programs, including access to essential health care and education, as well as subsidies that reduce costs such as for fuel, electricity, or food. For example, in 2012, the ILO adopted a recommendation on Social Protection Floors (No. 202), described in detail below, that include universal access to health care and basic income security; it should be noted that these floors are distinct from IMF social spending floors.

Target 1.3 of the Sustainable Development Goals (SDGs) calls on countries to “implement nationally appropriate social protection systems for all, including floors, and by 2030 achieve substantial coverage of the poor and the vulnerable.” The IMF’s approach to social protection is less categorical. On the one hand, it has increasingly emphasized the importance of social protection over the past decade, reflecting its broader evolution on social issues.[38] Following its 2019 social spending strategy, it developed a Technical Note on social safety nets, which are the central focus of its emphasis on social protection.[39] The focus on social safety nets already reduces social protection to what are generally poverty-targeted programs, which the Note defines as typically “noncontributory transfer programs designed to protect households from poverty and destitution by ensuring some minimum level of economic well-being.”[40] It expects that IMF engagement on safety nets “is likely to continue to expand as member countries respond to a rapidly changing economic and fiscal landscape.”[41] At the same time, it has maintained some distance from the idea that social protection falls within its remit, and frequently emphasizes its limited expertise on these issues. For example, both the 2019 strategy and the Technical Note instruct staff to rely on development institutions’ expertise and the World Bank in particular.[42]

However, even where other institutions or governments design social protection programs, the Fund exercises enormous influence over governments’ capacity to invest in such protection, such as by setting fiscal consolidation (that is, revenue increases or spending restraints for debt and deficit reduction) targets and social spending floors. While these floors, which are described further below, are not meant to be the maximum amount a government can spend, in practice the fiscal targets leave governments little flexibility to spend more. A review by Oxfam of 15 loans approved in 2020 and 2021 found that nine governments’ actual spending was within 10 percent of the cumulative floor and four spent less than the floor.[43] Moreover, many IMF-mandated reforms directly impact programs that function to reduce poverty and inequality, such as reducing or eliminating subsidies, pension reform, and freezing or reducing public sector jobs or salaries.[44] It often explicitly relies on a combination of social spending floors and specific improvements in social protection programs—typically establishing or expanding targeted cash transfer programs—to “mitigate” the impacts of these measures. For example, in Egypt, where IMF programs led to removal of fuel subsidies and other major changes in economic policy, the staff report notes that, “To mitigate the impact of the reforms on the poor, the authorities intend to use part of the fiscal savings to strengthen the social safety nets,” including two World Bank-funded targeted cash transfer programs.[45]

In addition, the IMF’s suggestion that its staff should rely on the World Bank’s expertise on social protection is in tension with its own resistance to following the Bank’s lead in embracing, at least in principle, universal social protection over targeted schemes.[46] While the Bank is far from implementing this commitment in practice, the Fund, as detailed below, has not taken the step of making universal social protection a goal and continues to be an active champion of targeted social assistance.[47] In some cases, the Fund explicitly calls on governments to more narrowly target its social assistance, but even where it does not, its position has important implications for where it sets social protection floors.[48]

Targeted versus Universal Social Protection

There are diverging views as to whether social protection programs should be universal (also called categorical), which is to say accessible to an entire population or specific categories of people (for example, children, older people, or people with disabilities) regardless of economic status, or whether it should be targeted based on people’s income, assets or other criteria such as the type of housing they live in (that is, means-tested). The IMF has largely promoted means-testing and in some countries, it has explicitly called for replacing universal programs with targeted ones.[49]

The appeal of a targeted approach seems intuitive: in a context where resources are limited, directing funds to those most in need appears to make sense. However, under human rights law, all people have the right to social security. Moreover, there is a robust body of research documenting that targeted programs, as a result of both their design and implementation, are expensive, prone to errors, can exacerbate social divisions and mistrust in government, and fail to adequately protect rights.[50] A 2017 study by the ILO and Development Pathways, a consultancy group that supports social protection systems in developing countries, found that targeting methodologies often result in errors due to both their design and implementation. For instance, targeting often relies on dated surveys, despite the reality that household composition, income, and needs are highly dynamic.[51] It also found that targeting that leads to arbitrary exclusion can generate social conflict and divisions within communities.

In recent years, a range of international voices, including the ILO and World Bank, have embraced universal schemes. In 2015, the World Bank announced it would partner with the ILO to promote universal social protection and in 2019 it published a white paper, Protecting All, that “wholly endorses the objective of universal social protection espoused by the international development community.”[52] The World Bank also joined numerous bodies, including the African Union and European Union, to form USP 2030, an initiative it co-chairs to achieve universal social protection by 2030.[53] Recognizing resource constraints, some economists, including in the 2019 World Bank white paper, have advocated for “progressive universalism,” wherein states begin by supporting the poorest segment of the population and gradually expand the program to include all. While a large gap between the World Bank’s position and its project designs remains, the Bank’s rhetorical embrace of universal social protection is significant.

The IMF, in contrast, has remained wary about universal social protection even as it recognizes, particularly in recent years, the shift in international consensus. For example, in December 2018, the IMF’s Finance & Development Magazine published several articles that strongly argued for the IMF to embrace universal social protection in line with emerging economic research and international practice.[54]

More significantly, in a new strategy on social spending approved in 2019, discussed more below, the IMF Board sought to clarify its position on targeted and universal schemes. The strategy acknowledges that in the majority of cases, mission chiefs recommend means-tested social protection, and in nearly one in five cases they recommend downsizing programs that are not means-tested.[55]

At the same time, recognizing the ILO and World Bank’s position on universal protection, as well as the approach of progressive universalism, the strategy asks mission chiefs to take a flexible approach, taking into consideration “country preferences and circumstances, including administrative, financing, social, and political constraints.”[56] While this opens the door to more frequently pursuing a universal approach, the language is cautious and doesn’t explicitly question mission chiefs’ overwhelming preference for targeted schemes to date.[57] Moreover, the Technical Note on social safety nets appears to double down on targeting and explicitly favors conditional cash transfers over other forms of social protection.[58] Although, like the social spending strategy, it directs staff to take into consideration “government objectives” when determining whether to target, and heavily emphasizes safety nets as a tool for reducing poverty.[59] In such cases, it recommends limiting transfers to only those living under the poverty line in amounts that fill the gap between household income and the poverty line, an approach that is reflected in several recent IMF programs that rely heavily on targeted cash transfer programs to mitigate the impact of IMF-driven reforms.[60] Any transfers to households whose income is above the poverty line, or amounts beyond what is needed to support households to reach the poverty line, are considered inefficient “leakage.”

At the same time, the note acknowledges many of the well-known problems with targeting. It highlights, for example, the risk of exclusion errors, noting that the “the capacity to effectively verify and monitor beneficiary incomes and other eligibility criteria … can be especially demanding in low-income settings where, for example, informal employment is prevalent and population registries are weak.”[61] It also acknowledges that “innovative approaches that rely on new technologies” could have “important trade-offs,” allowing for faster expansion though information they rely on may be “less accurate.” One way to address these challenges the note proposes is, rather than focusing on targeting the benefits, to “enhance tax capacity and effort” to “claw back” some of the benefits that accrue to high-income households.

Social Protection Floors

The 2009 G-20, meeting under the long shadow of the 2008 Financial Recession in Pittsburgh, Pennsylvania, marked an advance for international recognition of the importance of social protection. The concluding statement of labor and employment ministers recognized that their countries have a “shared responsibility” to “where needed, improve social safety nets,” and directed ministers to review reports from the ILO and other organizations to determine whether further measures were needed to address the job crisis, including “related to social protection programs.”[62] Building on this momentum, countries requested the ILO to find a common position agreeable to all governments, federated workers, and employers during its June 2011 conference calling on states to establish social protection floors with a goal of providing “universal coverage of at least minimal levels of protection.”[63] This call for social protection floors, echoed by others, was endorsed by the G-20 the same year, which defined it as “comprising a basic set of social guarantees for all and the gradual implementation of higher standards.”[64]

In 2012, delegates representing governments, workers, and employers at the International Labour Conference approved Recommendation 202 on Social Protection Floors, a new international labor standard that calls on member states to establish social protection floors, based on the principle of universality, that guarantee access to health care and provide basic income security for children; sickness, disability, or unemployment; and older persons.[65] As part of the UN Sustainable Development Goals (SDGs), adopted in 2015, countries committed to implementing nationally appropriate social protection systems for all, including social protection floors to reduce and prevent poverty.

IMF Conditionalities: Quantitative Performance Criteria (QPC), also simply referred to as performance criteria (PC): Specific, measurable conditions for IMF lending that require an IMF Board waiver if unmet. Many are quantatita. Indicative Targets (IT): Flexible numerical trackers for quantitative indicators to help monitor progress in meeting a program’s objectives. Structural Benchmarks (SB): reform measures that often cannot be quantified but are used as markers to assess program implementation: steps a country agrees to take before the IMF approves financing or completes a review.

Around this time, the IMF began to integrate social spending floors into its lending; a concept distinct from social protection floors defined differently in each program but frequently encompassing a broad range of public spending. In 2009, it introduced the Poverty Reduction and Growth Trust (PRGT), a facility for low-income countries that focuses more attention on social and other priority spending, including through incorporating explicit “targets” (typically social spending floors) into program design.[66] These targets, also called quantitative conditionalities, specify an amount of spending, either nominal or relative to an indicator such as GDP. They are incorporated into programs either as Performance Criteria (PC), which require a Board-approved waiver if not met in order to receive additional tranches of funding, or as indicative targets (IT), which staff have discretion to waive if unmet.[67]

However, despite the name, these floors were largely not specific to social protection.[68] For example, a 2017 review by the IMF’s Independent Evaluation Office (IEO), which conducts independent evaluations of IMF policies and activities, analyzed the IMF’s efforts around social protection and found that 55 of the 57 IMF arrangements approved under the PRGT between 2010 and 2015 had social or other priority spending floors, but only 10 (in seven countries) contained indicative targets defined to focus primarily and specifically on social protection.[69] Moreover, multiple reports from the Independent Evaluation Office of the IMF found that “Social and other priority spending floors in IMF-supported [Low-Income Country] programs were generally not very useful for safeguarding social protection expenditures.”[70] For this reason, this report refers to such floors in IMF programs as “social spending floors.”

Beginning in 2011, the IMF and the ILO started a limited collaboration on social protection focusing on a few pilot countries. The 2017 IEO review described the collaboration as “not very successful,” noting that of the three countries to pilot the initiative, two never appeared to take off and the success of the third was unclear.[71] In 2014, the IMF announced plans to scale up the program, but the review found that it was abandoned for reasons staff were unable to explain. The IMF has since relied mostly on the World Bank for expertise on social protection. In 2021, the IMF and ILO began collaborating on a new pilot program that included four countries, only one of which, Mozambique, was a loan program.[72] The pilot ended but Helmut Schwartzer, who heads the ILO’s Public Finance, Actuarial Services, and Statistics Unit in the Social Protection Division, described the engagement as “relatively positive” and told Human Rights Watch that they are working on a proposal “to continue and deepen the relationship over the next two years.”[73]

The IEO review, which included a host of recommendations, spurred the IMF Board to adopt a new Strategy on IMF Engagement on Social Spending in 2019.[74] The strategy recognized that social protection spending is important for inclusive growth and directly decreases income inequality. It provides more detailed guidance on when social spending is macro-critical, and therefore a permissible focus for IMF staff, breaking it down into three categories: whether social spending is sustainably financed, adequate, or efficient. Macro-criticality alone, however, is “necessary but not sufficient,” and staff are expected to continue to rely on external institutions where the IMF lacks in-house expertise.

At the same time, the strategy recommends strengthening quantitative targets and structural benchmarks to improve social safety nets, particularly when adjustments in Fund programs affect people in poverty. The strategy notes that “programs should aim to mitigate the adverse effects of adjustment measures on the vulnerable, and conditionality should support social objectives where critical for the program’s success.” Specifically, staff are expected to “analyze and, as appropriate, document the social impact of adjustment and measures to protect the vulnerable,” noting that “this may be important in order to build public support and political ownership of the program.”

The Technical Note on Social Safety Nets specifies the role of social protection schemes in achieving this goal:

In some cases, SSN reforms (for example, expanding existing cash transfer programs) are needed to mitigate the adverse impact of needed macroeconomic adjustment on vulnerable households, and therefore instrumental in supporting program success and to maintaining reform momentum.[75]

Where social safety net reforms are critical to IMF programs, the Technical Note allows for conditionalities either on general spending or on coverage amounts, and specifies when staff should use Performance Criteria or indicative targets to achieve these goals.[76]

Rhetoric versus Reality

The IMF’s focus on social spending and social protection has evolved significantly since the 1990s, yet both internal and academic research has cast doubt on how effectively it has been incorporated into the Fund’s operations or changed its impacts. The Fund’s own analysis of all 283 programs approved and ongoing between 2000 and 2016 found a “significant number of instances in which education and health spending decreased in program countries,” particularly where short-term fiscal consolidation is high.[77] And the IEO’s 2017 review of social protection found that although there has been an increase in attention to social protection, implementation has been uneven: “In some cases, staff provided high-quality analysis, but at times it seemed that attention to social protection in surveillance devolved into a box-ticking exercise.... Country officials noted that often advice was generic and lacked appreciation of country circumstances.” While programs “invariably emphasized the need to mitigate potential adverse effects of program measures on the most vulnerable,” authorities had mixed views on the effectiveness of IMF advice.[78]

Academic research using both quantitative and qualitative methods to assess changes in IMF programs over time has similarly found reason to doubt the practical impact of the evolution in the Fund’s rhetoric. One 2016 study by Alexander Kentikelenis published in the Review of International Political Economy analyzed all conditions in IMF loan agreements between 1985 and 2014, spanning 131 countries, and found “little evidence of a fundamental transformation of IMF conditionality”:

The organization's post-2008 programmes reincorporated many of the mandated reforms that the organization claims to no longer advocate and the number of conditions has been increasing. We also find that policies introduced to ameliorate the social consequences of IMF macroeconomic advice have been inadequately incorporated into programme design.[79]

A working paper by the Global Development Policy Center at Boston University analyzing the level of fiscal consolidation required in each IMF program from 2008 to 2018 similarly found that “IMF austerity did not significantly change in the wake of the financial crisis.” Strikingly, it found that countries that were granted relatively more relaxed fiscal conditionality were those with “larger voting rights in the IMF” or were politically aligned with Western Europe.[80] An in-depth analysis by Osama Diab in Middle East Critique comparing the IMF’s conditions in two major loans to Egypt in 1991 and 2016 also did not find evidence of a transformation in the IMF’s approach.[81] It found that the social component of the 2016 loan was roughly equivalent to the social component in the 1991 loan.

Yet another research paper, published in 2013 under the auspices of the UN Department of Economic and Social Affairs with research collaboration of IMF staff, examined episodes of fiscal consolidation measures for a sample of 17 countries that are members of the Organisation for Economic Co-operation and Development (OECD) over the period 1978 to 2009 and found that these episodes “typically led to a significant and long-lasting increase in inequality” and to a “fall in wage income,” but not in profit or rent income.[82] A 2020 study in the Review of International Organizations, found that IMF programs led to increased inequality, with the effect persisting for up to five years, although it wasn’t examining whether programs changed with IMF rhetoric and looked at programs between 1973 and 2013. Interestingly, it found that “this effect is driven by absolute income losses for the poor and not by income gains for the rich,” and that the evidence indicates that “inequality rises faster during programs that feature more extensive conditionality and that include social-spending cuts and labor-market conditions.”[83]

While these studies highlight the importance of closely scrutinizing how effectively the IMF is integrating its recent commitments to social protection, that does not mean the change in rhetoric is not itself significant. In September 2011, a provocative article appeared in the Cambridge Journal of Economics called “Not Your Grandfather’s IMF: Global Crisis, Productive Incoherence, and Development Policy Space,” in which the author, Ilene Grabel, made the case that the IMF’s critics who saw the institution’s response during the Financial Recession as fundamentally unchanged are missing an important point.[84] Rather than “strong continuity,” she argued that the IMF’s response to the global financial crisis was one of “productive incoherence,” which is to say the absence of a “unified, consistent and universally applicable approach” that had previously characterized its response to crises. In Grabel’s view, this inconsistency signals a weakening of the straitjacket neoliberalism held over policy space and opens significant, if fragile, possibility for change. A decade later, both views have reason to claim vindication: the research suggests the IMF’s impact does not reflect the transformation of its rhetoric but at the same time the “productive incoherence” has only deepened.

Less than a year after the IMF adopted a new strategy on social spending, the Covid-19 pandemic sent the global economy into a tailspin, causing one of the worst economic crises in living memory. From the outset, IMF Managing Director Kristalina Georgieva promised to use Fund resources to turn the recovery from the crisis into an opportunity to “build back better,” including by making economies more inclusive and less unequal. For the first year or so of the pandemic, the IMF mostly relied on emergency facilities that do not have conditions and therefore offer little room to address structural or long-term gaps in social spending. As the crisis has worn on, the IMF is increasingly turning to multiyear programs that include conditionalities, testing the IMF’s willingness to make its rhetoric a reality.

 

II. Social Spending Floors in the Wake of Covid-19

If the 2008 Financial Recession put social protection on the global agenda, the Covid-19 pandemic exposed how little progress was made in achieving that goal. The ILO estimates that around half the world’s population—more than 4 billion people—are not covered by any form of social protection.[85] And access is deeply unequal across countries: high-income countries spend an average of 16.4 percent of GDP on social protection whereas low-income countries, which are more likely to seek IMF support, spend an average of just 1.1 percent, according to the ILO.[86] The vast majority of those who are covered receive amounts too meager to protect their economic, social and cultural rights, and many countries’ social protection programs are mired in dysfunction or corruption such that benefits never reach those who are eligible.[87]

During normal times, this yawning coverage gap leaves people struggling to pay for essentials such as food or rent due to low wages or inability to work; as governments ordered lockdowns and other restrictions to prevent the spread of the Covid-19 virus, it turned catastrophic. Many governments, including in developing countries, introduced an unprecedented expansion of social protection benefits as well as other measures to support people affected by the pandemic, with nearly 1,700 measures reported in 209 countries between February 2020 and May 2021.[88] Of these measures, 52 percent were new emergency benefits, and the remaining measures were adjustments to preexisting programs.[89] However, the inequalities in social protection between countries are vast, and many simply did not have the infrastructure in place to reliably expand measures.[90] Many countries ended these expansions or programs even as people struggled to cope with the continued economic impacts of the pandemic and subsequent events, like Russia’s invasion of Ukraine. As a result, untold millions of people around the world were left without help when their income disappeared.[91] The World Bank estimates that 97 million people were pushed into poverty in 2020 alone.[92]

The scale of worsening poverty was not only devastating to people’s rights but also intensified the immense pressure the pandemic placed on economies, making it harder for them to recover. Indeed, strong pre-existing social protection systems contributed to making countries more resilient to the economic shocks of the pandemic.[93] In the United States, where the government responded with an unprecedented expansion of benefits including near-universal cash transfers, poverty decreased despite the economic toll of the pandemic.[94]

The IMF recognized from the outset of the pandemic the importance of getting support to people and repeatedly exhorted governments to “spend what they can” to protect “lives and livelihoods.”[95] To support governments’ response, between March 2020 and March 2022, the Fund provided $168 billion through its lending facilities and debt relief to upwards of 100 countries.[96] In August 2021, the Fund also approved a new allocation of $650 billion in Special Drawing Rights, a Fund-created interest-bearing asset that can be exchanged for certain currencies and is distributed to all member countries in proportion to their quota.[97]

But particularly for countries without existing or reliable social protection programs, often compounded by other gaps such as in health and education systems, the infusion of resources did little to protect people or bolster economic resilience. The pandemic made clear that governments need to invest in adequately funded permanent programs, both to build more equitable and stable economies in the aftermath of the pandemic and, as the IMF notes, to prepare for a world with increasingly frequent economic shocks due to climate change, technological advancements, financial crises, and other drivers.[98] For example, an IMF departmental paper published in 2020 focusing on the Middle East and Central Asia finds that a 10 percent increase in public social protection spending per capita sustained for three years can close 20-40 percent of the Human Development Index (HDI) gap between countries in the region and their comparators, and up to 45-60 percent of the gap could be closed if this increase in social spending is accompanied by better governance.[99]

At the same time, governments, particularly in low- and middle-income countries, faced reduced revenues and mounting debt as a result of the pandemic. As the IMF transitioned from relying principally on emergency loans, which do not include conditionalities, to support governments’ immediate pandemic response, to negotiating multiyear programs to support economic recoveries, it faces a critical opportunity to break with its historical reliance on fiscal consolidation and structural adjustments that fuel economic inequality and to promote adequate investment in social protection.

In this context, the IMF has emphasized social spending floors, which it also sometimes calls social protection floors. The former label is more accurate since many floors encompass categories of spending that go far beyond social protection. IMF loan documents describe these floors as simultaneously serving multiple distinct purposes: they protect certain spending from fiscal consolidation, mitigate the impact of adjustments in programs that negatively impact people on low incomes, and address longstanding gaps in social protection programs.

Doubling Down on Austerity

In addition to emergency relief loans and debt relief programs, between March 2020 and March 2023, the IMF approved 33 loans under the Extended Credit Facility (ECF) and Extended Fund Facility (EFF) and six under the Stand-By Arrangement (SBA) with a total amount of $30 billion to the following countries: Afghanistan (frozen), Argentina, Armenia, Bangladesh, Barbados, Benin, Cabo Verde, Cameroon, Chad, Congo, Costa Rica, Democratic Republic of Congo, Ecuador, Egypt, Gabon, Gambia, Georgia, Guinea-Bissau, Jordan, Kenya, Madagascar, Malawi, Mauritania, Moldova, Mozambique, Nepal, Niger, Papua New Guinea, Senegal, Serbia, Seychelles, Somalia, Suriname, Tanzania, Uganda, Ukraine, and Zambia.[100] The total number of people living in these countries is over 1 billion.

All three are lending instruments that include conditionalities: ECF and EFF are generally three- to five-year programs, and the SBA generally extends for 12 to 36 months. The conditionalities included in loan programs are the result of often intense negotiation between IMF country teams and government authorities and are therefore to some extent shaped by the particularities of each country’s economic reality as well as the individual views and preferences of authorities and staff involved in these discussions. IMF guidelines also encourage staff to engage civil society throughout the negotiation process.[101]

Moreover, loan agreements, and the policies that underpin them, are shaped not only by staff, but by powerful governments on the IMF’s Executive Board, which have to approve them. The Board is made up of 24 directors who are elected by member countries or groups of countries, and the managing director, who serves as the chairperson.[102] Voting power on the Board is distributed roughly based on countries’ economy size, such that the US and European countries, as well as Japan and China, have outsized power including on matters related to the institution’s economic approach.

This process means that each loan agreement is different, and governments, and sometimes the public, have a hand in shaping them. This report does not examine the specific context that led to every program, nor does it seek to make determinations about individual policy choices they reflect. Rather, it traces some of the core elements and underlying approach that thread through these loans to reflect the Fund’s institutional philosophy embedded formally through policies and guidance as well as informally through high-level speeches and inter-departmental discussions.

The 39 loan programs reviewed vary in many respects but nearly all share a common theme: they preserve the IMF’s traditional fiscal consolidation measures while increasing the emphasis on “mitigating” the impact of IMF-driven reforms. A table with key information about each program is in the following section and a more detailed summary is included in an annex to this report. All but six programs explicitly impose fiscal consolidation and nearly all of these include a variation of the phrase “while protecting social spending.” Even those programs that don’t explicitly reference fiscal consolidation may nonetheless reduce public spending in ways that risk harming rights. All but six programs include a social spending floor.[103] The IMF had already been moving in the direction of emphasizing social spending floors in recent years, so it is not entirely new, but it is notable that the sense of urgency to build more equitable and inclusive economies triggered by the pandemic has led the IMF to double down on this approach.[104]

The potential risks that fiscal consolidation, which simply means reducing public debt or deficits, poses for human rights depend on the specific policies governments pursue to achieve consolidation targets. It is possible to achieve these targets by growing GDP or raising new revenues through progressive means, and some programs emphasize a “revenue-based” approach or a focus on growth.[105] However, this report finds that most of the IMF programs approved since the start of the pandemic project decreases in public spending, and 31 of the 38 reviewed include at least one measure that risks harming rights, as discussed in detail in the following section.[106]

International Human Rights Standards and Austerity Austerity has well documented impacts on human rights. International legal standards have evolved over the past decade to ensure that rights are protected during economic crises, as detailed in Section IV of this report. In 2016, the UN Committee on Economic, Social and Cultural Rights (CESCR), charged with interpreting the ICESCR, issued a statement on public debt and austerity that made clear that international financial institutions (IFIs) “are obligated to comply with human rights” and explicitly disagrees with the position that the IMF’s Articles of Agreement preclude it from doing so.[107] In March 2019, the Human Rights Council adopted a set of principles, developed by Juan Pablo Bohoslavsky, the UN Independent Expert on foreign debt, to guide states and financial institutions in their response to economic crises.[108] These principles incorporate and update previous related standards and provide that: IFIs and governments have human rights obligations to respond to economic crises in ways that protect and advance rights in both the short and long term by pursuing economic recovery measures that further “the benefit of the whole population, instead of only a few.” Debt servicing should not come at the expense of rights and creditors should consider debt restructuring or relief where appropriate to ensure governments are able to adequately protect rights. Fiscal policies, including in relation to revenue collection, budget allocations and expenditure, must comply with international human rights standards and principles, in particular equality and non-discrimination. Any measure that would result in retrogression—that is, reversal or erosion—of rights is “considered a prima facie violation” of rights, and is only permissible if there are no other policy alternatives that would be less harmful to rights. To pursue austerity measures, states must “demonstrate that all other alternatives have been exhausted, and provide safeguards to protect human rights, particularly the rights of the most vulnerable.” Where governments are able to demonstrate they are unable to avoid measures that lead to retrogression, such measures must be temporary, proportionate, non-discriminatory, and protect the minimum core of economic and social rights. They also must be based on transparency and the genuine participation of affected groups and subject to meaningful review and accountability procedures, including human rights impact assessments. As part of IFIs and governments’ obligation to engage in a meaningful public dialogue, particularly with those most likely to be affected, they are expected to conduct human rights impact assessments “that demonstrate that their proposed economic reform measures will realize, not undermine” human rights. The assessments, which should be conducted in a transparent, participatory process and then published, “should seek to identify and address the potential and cumulative impacts of measures on specific individuals and groups and protect them from such impacts.” In the case of IMF loan programs, they should be published prior to approval.

There is a significant body of economic research questioning the effectiveness of austerity and documenting its harms, alongside the evolution of international human rights standards. Nobel Prize laureates Joseph Stiglitz and Paul Krugman have criticized austerity as an ineffective and harmful response to crises.[109] As noted in the Background section, the IMF’s chief economist issued a report that found that fiscal consolidation led to economic contraction in developed countries, whereas the IMF had predicted it would lead to economic growth, and a study of fiscal consolidation in 17 OECD countries found it typically led to economic contraction, falling wages, and long-term inequality.

Numerous economists urged IFIs and governments to find alternatives to austerity in their response to the economic crises precipitated by the Covid-19 pandemic. In January 2021, the Chief Economist at the OECD, Laurence Boone, told the Financial Times, “The first lesson [of the financial recession] is to make sure governments are not tightening in the one to two years following the trough of GDP.”[110] In September 2020, the UN Conference on Trade and Development warned of a “lost decade” if countries choose the path of austerity.[111]

While in the early months of the pandemic the IMF urged countries to spend what they needed and appeared to eschew austerity as a path to recovery, it quickly pivoted to pushing countries toward fiscal consolidation.[112] In the first year of the crisis, the Fund approved emergency loans to dozens of governments that are lump-sum up-front payments, and therefore don’t include conditionalities, although governments’ request letters include specific commitments that reflect and reinforce policy plans.

A review by Oxfam found that 85 percent of the 107 Covid-19 loans included commitments to impose fiscal consolidation as early as 2021.[113] Isabel Ortiz and Matthew Cummins published a paper in September 2022 analyzing IMF policy advice for 172 countries and similarly found that most included austerity-driven measures. Although the paper looked at advice rather than loan conditionalities, it is nonetheless influential and indicative of the IMF’s approach. For example, it found 120 governments were considering “rationalizing” social protection programs, which often means more narrowly targeting them. This pivot, which appears to mimic the Fund’s abrupt shift from stimulus to austerity following the Financial Recession, poses significant risks to human rights. These risks are greatly exacerbated by the absence of impact assessments or justifications for pursuing these policies, which in turn undermines robust national dialogue in arriving at these policy choices.

The IMF’s World Economic Outlook published in April 2023 cites findings, including by internal IMF research, that yet again call into question its heavy emphasis on fiscal consolidation and cutting public spending. The report confirms, for example, that “fiscal consolidations do not reduce debt ratios, on average.”[114]But, as described below, this continues to be the IMF’s approach to the current crisis.

Human Rights Risks of Austerity-Driven Measures in IMF Programs

This report tracks three common reform measures to achieve fiscal consolidations targets that have been shown to often negatively impact people on low incomes: capping or reducing public wage bills; increasing value-added taxes or removing exemptions; and reducing or removing subsidies. While this report looks only at IMF program conditionalities, the analysis conducted by Isabel Ortiz and Matthew Cummings of 285 country reports, which also include Fund advice outside of programs, found these were among the most common measures included to achieve fiscal consolidation: 91 countries were considering public wage cuts or caps and 80 were considering reducing or eliminating subsidies, predominantly on energy, food, and agricultural inputs.[115] It also found that 120 governments were considering rationalizing or targeting social protection systems, which often indicates cuts.[116]

This report’s analysis is not intended to be a comprehensive review of each program. Other measures that can negatively affect people on low incomes, such as pension reform, labor flexibilization, and changes to exchange rates, are not addressed in this report. The analysis also does not include reforms that may positively impact rights, such as measures related to anti-corruption, transparency, and good governance, although this issue receives notable attention in several programs. For example, improving public procurement and bringing transparency to State-Owned Enterprises is a major focus of the program in The Gambia.[117] Several loans, including to The Gambia, also require governments to implement commitments to audit and publish information regarding their Covid-19 spending made under emergency loans.[118] Moreover, several loans seek to achieve fiscal consolidation targets in part by raising revenues through progressive taxes and/or by reducing tax evasion by the wealthy and by corporations. However, even where present, they generally accompany, rather than replace, reforms with negative impacts that disproportionately fall on people on low incomes.

A table with the main findings is below, with a more detailed summary of each program included in an annex to this report. The analysis is largely limited to the IMF staff reports that accompany the Board’s approval of loan agreements. Every IMF loan is periodically reviewed by IMF staff before a new tranche of money is released and conditionalities are often updated to reflect unexpected events, among other reasons. Even where the terms of a program have not officially changed, some governments may not fully implement everything to which they’ve agreed, and it is possible that some of reforms below have yet to be implemented or never will be. By the same token, some programs do not include a tracked measure because the government already implemented it, in some cases under previous IMF programs. For example, Jordan increased VATs and removed exemptions in 2018 under an IMF program, so these measures do not appear in the program approved in 2020. Nevertheless, the analysis provides an important window into the IMF’s institutional approach to lending in the wake of the Covid-19 pandemic.

Table of IMF Programs: March 2020 – March 2023

Freezing Hires and Capping or Lowering Salaries of Public Sector Workers

21 of the 33 programs reviewed include broad structural benchmarks or general advice to lower the public wage bill, through freezing hires, which reduces the number of public sector jobs through attrition, and capping or cutting salaries. Even where nominal spending is maintained or increased, this can lead to an effective cut in salaries due to inflation or inadequate growth given demographic changes. For example, the Barbados loan notes that “public wages in Barbados have been kept constant in nominal terms since 2018; hence real wages have declined significantly, especially in 2022.”[119] Moreover, an analysis of 17 loan programs approved in 2020 and 2021 found that the IMF “appears to generate optimistic estimates and projections about the trajectory of wage spending at the beginning of a loan, and over time revises its projections downwards.”[120]

In some cases it may be appropriate for governments to reduce jobs in certain specific public sectors, or reduce the salaries of the highest earners, but experiences from past reductions in public wage bills have shown that such reforms can harm governments’ capacity to deliver quality public services and efficient administration in ways that harm rights.[121] A 2021 review by ActionAid, a non-governmental organization, of IMF programs in 15 countries from the last five years found that combined they blocked the recruitment of over 3 million nurses, teachers, and other public essential workers.[122] The report also found that there appeared to be no benchmarks, objective reference points, or systematic analyses that led to advice to cut public wages, nor was there a more context-specific consideration of past experiences with public wage cuts or freezes or existing staffing shortages, such as for teachers or nurses. As a percentage of GDP, spending on public wages was significantly above average in only one of the 15 countries, and 10 were below average. “Indeed, even countries who have exceptionally low spending on the public sector wage bill have still been advised to cut, rather than increase."[123]

Erosion of pay can also negatively affect staff performance or lead to an increase in demands for bribes.[124] For example, a UNICEF survey following public wage cuts in Cambodia found an increase in absenteeism and decrease in working hours.[125] Cuts and caps of employees or their wages working in health and education can compromise the delivery of these services and therefore pose a particular risk to rights, although several of the loans explicitly exempt one or both of these sectors from such measures.

The degree to which the IMF is relying on public wage bills and caps in its Covid-era lending is in contrast to a commitment it made in 2007 discouraging wage bill caps for low-income countries. It urged those who use them to justify doing so in reports and reassess the need and rationale in program reviews.[126] The decision followed an internal review that found “that despite their intended short-term nature, wage bill ceilings tended to persist in [Poverty Reduction and Growth Fund]-supported programs, were difficult to monitor, and distracted governments from efforts to strengthen institutions for better controlling wage spending.”[127] IMF Directors noted they “hoped that the use of medium-term expenditure frameworks and strengthened budget and payroll systems will gradually obviate the need for such ceilings,” but in the meantime they should be used only in exceptional circumstances and be flexible. In particular, they called for staff reports to “justify in a transparent manner the use of wage bill ceilings and for a reassessment of their need and rationale at the time of program reviews.”[128]

As with many IMF adjustment programs, cuts and caps to public sector jobs or wages have been found to disproportionately impact women.[129] First, “women and girls are more likely to fill the gap when public social services are absent,” as noted by a 2017 analysis by ActionAid on the gendered impact of IMF policies.[130] In addition, women tend to be overrepresented in public sector jobs.[131] An IMF guidance note on how to operationalize gender equality in country work acknowledges this risk and specifically calls on IMF staff to consider alternatives:

Some policies recommended by staff to support growth and stability may have differential gender impact that could exacerbate gender inequality. For example, budget cuts on subsidies and social programs, cuts in the public-sector wage bill, or increasing transportation fees might have larger impact on women. In these instances, staff may consider an alternative policy mix to prevent such negative externalities or—if the former is not feasible—suggest some mitigating measures.[132]

Under international human rights law, economic policies cannot be discriminatory in their impact on rights whether directly or indirectly.[133] The guiding principles developed by the UN expert on foreign debt specify the risk of public wage cuts exacerbating gender inequalities:

Therefore, economic reforms which encourage, among other things, labour market flexibilization, reductions in the coverage of social protection benefits and services, cuts to public sector jobs and the privatization of services tend to have a negative impact on women’s enjoyment of human rights.[134]

IMF staff assessments should include the impact on women and girls, older persons, people with disabilities, and other groups who may be disproportionately harmed by specific policies. As discussed more in the following sections, of the 33 programs reviewed, only 2 included IMF staff assessments of the impact of public wage cuts on poverty and inequality. However, both analyses are hampered by an extremely narrow view of these issues such that reductions in public wages are seen as positive. The staff report for the program in Costa Rica limited its analysis to whether lowering wage bills would increase poverty and concluded: “that a reduction in the public sector wage bill, reflecting the authorities’ plans to freeze base salaries and accrual of annuities, and elimination of all vacancies, among other measures, would not have an impact on the poverty rate as public sector employment is concentrated in the higher deciles of the income distribution.”[135] An analysis of the reduction of the public wage bill in a program in Ecuador likened cutting wages to progressive taxation, arguing: “Since public-sector employees earn above the median income” reductions in wage bill “is equivalent to a progressive income tax and would go towards reducing income inequality.”[136] But combatting inequality should be achieved by ensuring everyone earns at least a living wage, ensuring everyone can realize their economic and social rights, and reversing the extreme concentration of wealth, not decreasing middle-income wages and jobs. Moreover, these analyses don’t take into account the impact of any erosion in access or quality of public services due to these measures.

Cuts or Reductions of Subsidies to Fuel and Electricity

Sixteen of the reviewed loan programs call for removing or reducing consumer-based subsidies for fuel and electricity or developing a plan to do so. Many of these programs were approved after governments increased subsidies to help people cope with the increase in fuel costs following Russia’s invasion of Ukraine. From a human rights perspective, removing or reducing fossil fuel subsidies is distinct from other measures tracked in this report, since they underpin an economic model that fuels inequality and climate change and should therefore be phased out to realize rights. But doing so without taking adequate measures to safeguard rights and invest in transition to renewable energy can significantly raise the cost of living and negatively impact people on low incomes (see Textbox).

Shifting Social Contracts: Fossil Fuel Subsidy Removal, Climate Change, and Human Rights Many governments subsidize the cost of fuel or electricity for consumers. In 2022, governments spent $1 trillion on fossil fuel consumer subsidies, according to the International Energy Agency, double the previous year due to higher prices triggered by Russia’s full-scale invasion of Ukraine.[137] These subsidies help people on low incomes to cope with short-term price spikes, but they also contribute to the climate crisis and thereby represent significant additional risks to the very same populations whose rights are most affected by fuel price hikes. Fossil fuel subsidies form a fundamental part of the social contract in many countries, particularly when energy costs rise as they did in 2022. This is particularly true in the Middle East and North Africa, which is responsible for almost half of global energy consumption subsidies and has historically invested very little in social protection.[138] Increasing fuel and electricity prices can drive up prices more broadly, particularly for transportation, housing, and utilities—which make up 70 percent of household expenditures in developing economies.[139] IMF research found that the direct and indirect impact of a $0.25 per liter increase in fuel prices results in an average 5.5 percent decline in household real incomes.[140] Fuel and electricity price increases are felt most acutely by people on low incomes who will pay a higher share of their income for goods or services that are essential to their rights.[141] At the same time, fossil fuel subsidies place an enormous burden on government budgets. For example, a staggering 40 percent of Lebanon’s debt is due to transfers to its national electricity company, contributing to its severe economic crisis.[142] These public resources are disproportionately captured by the wealthy, fueling inequality at the expense of funding programs essential to human rights, including social protection and public services.[143] By delaying the transition to renewable energy, they also contribute to climate change, which exacerbates inequality and disproportionately harms both individuals and countries without adequate resources to respond to the effects of climate change, such as extreme weather and natural disasters, threats to water sources and crops, and higher food prices. Phasing out fossil fuel subsidies is therefore an environmental, human rights, and economic imperative. If done in a way that advances, rather than erodes, human rights, the enormous savings subsidy removal generates presents an opportunity for building a new social contract that is aligned with human rights and environmentally sustainable. IMF research, while emphasizing the importance of removing subsidies for economic and environmental reasons, also acknowledges its complex social impacts, including that it can exacerbate poverty and inequality. In 2016, it developed a tool for staff to measure the distributional impact, but it is unclear if staff are conducting such analyses because, with rare exception, they are not published.[144] Moreover, the IMF’s position on removing fossil fuel subsidies is to “protect the poor through targeted cash or near-cash transfers,” which is far too narrow.[145] Rather, governments should use the savings to fund universal social protection programs and advance the transition to renewable energy by investing in renewable energy sources, as well as other investments critical to a green economy.[146] For example, accessible public transportation can help mitigate higher gas prices for those who previously relied on cars, which in turn further benefits public health and the environment.[147] In this way, phasing out subsidies can facilitate the shift to social contracts that better distribute resources to maximize all people’s ability to realize their economic, social, and cultural rights, including the right to social security. If its implementation is well sequenced and communicated, it can dissipate public opposition to subsidy removal, which has prevented or stalled fossil fuel subsidies removal, or prompted governments to restore at least some of them.[148] Investment in renewable energy not only addresses one of the major contributions to climate change, but also reduces the long-term cost of energy, rather than simply transferring the expense of generating energy from fossil fuels onto consumers. Recent modeling from the University of Oxford shows that an “accelerated zero-carbon deployment” would save the world $26 trillion in energy costs over the coming decades compared with continuing today’s fossil fuel-based energy system – including the costs to adapt the grid.[149] In short, fossil fuel subsidies are a relic of a social contract that fuels inequality and climate change. Removing them in a way that advances rights can help shift to a new eco-social contract that reduces inequality and advances the energy transition – two central aspects of the vision IMF Managing Director Kristalina Georgieva articulated at the start of the Covid-19 pandemic. But the IMF’s current approach of replacing these subsidies with targeted cash transfer programs, often long after the subsidies have been removed, falls far short of that vision and risks further eroding, rather than advancing, human rights.

Increasing Value-Added Taxes and Removing Exemptions

Around half of the loans analyzed include benchmarks or advice to increase value-added taxes (VAT) or remove VAT exemptions, such as on food. VAT is an “indirect” tax that is levied on goods and services that are consumed or sold, in contrast to “direct” taxes on personal or corporate income. Because such taxes apply the same rate to all people, regardless of their income, they tend to be considered regressive; that is, they weigh “proportionally more on lower income households since they consume a larger share of their income than richer ones.”[150] Where VAT is, for example, limited to luxury items, it may not be regressive, but that is rarely the case in the context of IMF programs; indeed, in many cases, program conditionalities remove exemptions on food and other staples. Some contend that VAT is not a regressive tax because, in absolute terms, higher-income households make up a higher share of the total revenue raised because they consume more.[151] Others consider VAT to be regressive but make the case that it is nevertheless necessary or that its regressive nature could be adequately mitigated via social protection programs.[152]

However, these debates are largely over the definition of regressivity rather than the impact of VAT since even those who believe VAT is not regressive generally agree that it burdens people on low incomes.[153] The Cameroon program, for example, singles out VAT exemptions on food and nonalcoholic beverages as disproportionately benefiting the wealthy before committing authorities to “phasing out the least effective measures and replacing them by conditional cash transfer mechanisms, with World Bank support.”[154] In a similar vein, a review of VAT in 27 OECD countries that argued it is generally not regressive under how it defined the term nevertheless concluded that “even a roughly proportional VAT can still have significant equity implications for the poor – potentially pushing some households into poverty.”[155]

Critically, VAT can also make food and other goods and services essential to rights no longer accessible to some people. In many countries, menstrual health products are subject to VAT, which can impede girls from attending school, among other harms.[156] In France, the poorest households pay more than 20 percent of their income in consumption taxes, compared to less than 10 percent for the wealthiest households.[157] A 2016 report by the Egyptian Initiative for Personal Rights, an independent human rights organization, found that the poorest 10 percent of Egyptians spend 6.4 percent of their incomes on a VAT introduced as part of an IMF program, nearly twice as much as the country’s richest, who spend 3.3 percent.[158]

In Sri Lanka, where an economic crisis was triggered in large part by having one of the lowest tax-to-GDP ratios in the world, the program heavily relies on VAT to increase revenues. In anticipation of the program, the government increased VAT from 8 to 15 percent, and by the end of the program 33 percent of all taxes are expected to come from VAT, far higher than the OECD average of 20 percent.[159] In Zambia, which has among the world’s highest poverty and malnutrition rates, the program specifies removing VAT exemptions on unprocessed foodstuff except “specific items that figure prominently in the food basket of the poor.”[160]

The IMF’s continued reliance on austerity in general, and these measures in particular, carry enormous risks of negatively impacting people on low income’s economic, social and cultural rights. As such, staff reports play an important role within the IMF review and approval process and should be aligned with the human rights obligations of the IMF and its member states. They should include an assessment in line with international human rights standards that explains why specific policies are most likely to lead to inclusive and sustainable economies, calculate any potential harmful impacts of conditionalities, and explain the steps taken to safeguard against them. These analyses should be both the product of and basis for continued engagement with the public in the country in question and should be made public prior to a loan’s approval. The IMF’s strategy on social spending, adopted in 2019, provides an avenue for staff to conduct such an analysis by directing staff to “analyze and, as appropriate, document the social impact of adjustment and measures to protect the vulnerable.”[161] However, as will be further discussed below, only 1 program of the 39 reviewed included a detailed assessment. Moreover, the continued reliance on a familiar mix of measures for reducing debt despite their acknowledged negative impact on people on low incomes puts a significant burden on social spending floors to differentiate the IMF’s response to this crisis from its austerity-driven approach following the Financial Recession. The adequacy and design of these floors is therefore of utmost importance.

Assessing Social Spending Floors

Lack of Adequate Information

Many IMF programs acknowledge the harmful impact of the adjustments they prescribe and call on governments to protect people from these impacts by improving social protection systems. To facilitate this, all but 7 of the 33 programs approved between March 2020 and March 2023—or 78 percent—include social spending floors. As described in the previous chapter, these floors are typically quantitative targets for government spending of specified social programs that serve multiple purposes, including increasing investment in social protection and other social spending, protecting certain social sectors and programs from austerity-driven budget cuts, and offsetting the costs of reforms that increase the cost of living. These objectives make social spending floors critical for protecting human rights, yet it is challenging to assess their adequacy for several reasons discussed below. Despite these limitations, available information and IMF policies, detailed in the following section, indicate that the floors are likely not adequate to protect rights.

The first challenge of assessing the adequacy of social spending floors stems from how they are defined in IMF programs. As is apparent from the table in the previous section of this report, the categories of spending each floor covers vary widely among programs, with some encompassing several ministries and others more narrowly defined to encompass specific social programs. For example, the social spending floor in the Seychelles largely includes programs typically referred to as social assistance, such as cash transfers for people in specific categories (people with disabilities, orphans, new mothers, etc.). In contrast, Benin’s program defines the floor as covering 72 enumerated programs in 12 sectors, including “integrated management of border areas” and “development of rural roads.”[162]

To some extent, this is to be expected, since each country has a unique set of needs and existing social protection programs and public services. But the range in definitions surfaces a tension between the multiple purposes these floors serve. Many are defined in ways that go far beyond social protection, or do not include such programs at all. Broadly defined social spending floors are critical for protecting the full range of economic, social and cultural rights, but at the same time it is difficult to assess the extent to which they effectively fully address the impact of structural adjustments or improve social protection, which, as discussed above, is in most cases an essential (though not only) component to ensure everyone’s rights are protected from austerity measures. The Benin program removed VAT exemptions on water, electricity, and imported rice, even as one in three children suffer from chronic malnutrition, according to the World Food Programme.[163] It also encourages the government to further increase the pump price of diesel and gasoline after having already increased it by cumulative 40 and 25 percent in 2022, respectively.[164] By defining the floor so broadly, it ensures minimum spending on Benin’s social sectors, but in the past, its social spending has been chronically low and there is no assurance that it will help struggling people offset the specific costs associated with the IMF program or improve Benin’s social protection system.

For this reason, this report recommends that the IMF adopt separate floors to serve distinct purposes: one for protecting and increasing spending in areas critical for fulfilling social and economic rights without discrimination, and another for mitigating against necessary measures that nonetheless pose risk to rights, such as phasing out fossil subsidies. The first should be broadly defined, and disaggregated by sector, and ensure that spending on, for example, health, education, and social protection are, at a minimum, in line with minimum international benchmarks as a percentage of GDP and of national budgets.[165] The second should be based on published assessments of program impacts and demonstrate how the target, such as increased social protection spending or coverage, adequately mitigates those impacts to avoid, at a minimum, the retrogression of rights. For both, the IMF should consider replacing “floors,” which are revised in each review, with “goals” to be achieved by the end of the program, and a plan developed from the outset to achieve specific benchmarks in each review.[166]

The IMF program in Uganda, approved in June 2021, and in Egypt, approved in December 2022, employ two separate social spending-related measures that are a step in the direction of this approach.[167] In the case of Uganda, one floor encompasses general social spending, covering all spending on health, education, and social development (excluding external financing) and a separate floor covers social assistance defined as several cash transfer programs. In the case of Egypt, a social spending floor covers the budgets of the ministries of health and social solidarity, while a structural benchmark requires the government to expand coverage of two cash transfer programs from 4 to 5 million.[168]

A second challenge is that, in most cases, it is not possible to assess whether, or to what extent, the social spending floor increases spending relative to previous years. Most loan agreements cobble together a motley of different budgets and programs into a new category not tracked in government budgets. So while the floors increase throughout the duration of the program, the baseline pre-program spending is often unclear. Where budgets separately track each component included in the floor, it may be possible to add these sums together to compare with the floor, but this can be both labor intensive and unreliable. And this is not possible when the included programs are not separately tracked in government budgets, or the budgets are not publicly available. For example, Jordan’s social spending floor is defined as including non-wage health and education current expenditures, the National Aid Fund’s social protection program, and school feeding programs. However, the budget does not include a specific line for “non-wage health and education current expenditures,” making it difficult to compare the amount provided for in the floor to spending prior to the program. Even where it is possible to do so, inflation and demographic changes may erode the benefits of increases. For this reason, baseline information should be included for all loans, and all nominal spending amounts should also be included as a percentage of GDP.

Finally, a central objective of social spending floors is to mitigate the social impact of program adjustments. As discussed, a mitigation-based approach is inherently problematic as it fails to demonstrate that the full set of policies pursued are the most protective of rights, but it is made even more so as the vast majority of programs do not even attempt to calculate social impacts despite guidance to do so in the IMF’s 2019 strategy on social spending. Nor do most programs include a detailed analysis of existing social protection programs. As a result, it is difficult to determine whether the social spending floor adequately mitigates the impact of these adjustments or will lead to a safety net that protects people’s right to social security. This key gap in information also impedes the public’s ability to understand policy choices in IMF programs and hold their governments accountable for these choices. All programs should include such an analysis and Staff Level Agreements should be published prior to Board approval to facilitate a fully informed public dialogue, engaging particularly those most affected by program adjustments, so as to ensure that the full suite of policies pursued are most protective of rights. The Ecuador program is the exception and includes a detailed analysis of both the costs of the adjustments and the expansion in cash transfers intended to offset them. This analysis, while instructive, also surfaces significant concerns with social spending floors, which are described below, and which are further reinforced by the IMF’s policies and Human Rights Watch research in Jordan.

Three Key Problems: Adequacy, Enforcement, and Sequencing

In October 2019, the Ecuadorian government under President Lenín Moreno formally withdrew from a program that the IMF had approved months earlier. The program pushed the government to adopt austerity measures,[169] triggering weeks-long protests led by Indigenous groups.[170] In particular, protesters objected to the removal of fuel subsidies at the heart of the program that had already led to spikes in gasoline and diesel prices that hit people on low incomes in rural areas particularly hard in the absence of adequate measures to mitigate these impacts. But the program also included other austerity-driven measures such as cuts to the public wage bill that resulted, for example, in the dismissal of 3,680 workers from the ministry of public health.[171]

A year later, in October 2020, the IMF approved a $6.5 billion Extended Fund Facility to replace the derailed program.[172] The new program maintained many of the features of the old one. Moreno’s administration began to reduce certain fuel subsidies even before the new agreement was reached, and under the program they committed to “rationalizing” the public wage bill through “employment reduction and lower salaries,” increasing the VAT by three percentage points and eliminating VAT refunds for older people, and reducing the income subject to personal income taxes to include an additional roughly 30 percent of the population.[173]

However, likely due to the fraught history, the new program includes two innovative features for “protecting the poor and most vulnerable” that are instructive.[174] First, Ecuador’s program is unusual in that its floor is defined not as a minimum amount of spending but the expansion of the cash transfer program to cover at least 80 percent of the poorest 30 percent of the population by the end of 2021.[175] Prior to the pandemic roughly 30 percent of the population—or 5.2 million people—lived under the poverty line, only 37 percent of whom were covered by any social assistance. Second, IMF staff conducted a detailed analysis to assess the program’s impact on poverty and inequality. The IMF staff’s analysis found that, because of the expanded safety net, the bottom three deciles (that is, the poorest 30 percent of Ecuadorians) would experience an effective increase in their income, while the rest of the population would experience a net decrease ranging from 2.1 percent (for those in the fourth decile) to 6.3 percent (for those in the tenth decile).[176] It found that the cash transfers would decrease inequality, although aspects of its analysis are problematic, as discussed in more detail below.

Ultimately, some of these initial conditionalities that were part of the analysis were not fully implemented. Following approval of the program, an election brought to power a new president, Guillermo Lasso, who implemented income and corporate tax reforms that increased taxes on the middle class but did not raise value-added taxes. In June 2022, Indigenous-led protests forced the government to freeze the removal of some fuel subsidies,[177] since their reforms had made the prices vulnerable to steep increases following Russia’s full-scale invasion of Ukraine.[178] However, the detailed poverty and inequality analysis remains instructive as a rare glimpse into the IMF’s approach to assessing the impacts of specific conditionalities and their efforts to mitigate them. While it is positive that the analysis took place, it surfaces three key problems with the IMF’s approach to social spending floors: adequacy, enforcement, and sequencing.

The Limitations of Poverty-Targeting

On paper, IMF policies direct staff to defer to “government social policy objectives” in assessing the adequacy and efficiency of social protection programs.[179] However, in practice, as discussed in the Background section, Fund advice and programs reflect a heavy preference for targeted cash transfer programs directed at households living beneath national poverty lines. Cash transfer programs can be inadequate in two ways: the number of people they cover and the benefit amounts they provide. By setting social safety net coverage, rather than spending, as a social spending floor, the IMF program in Ecuador offers a particularly useful example of the problems with this approach when assessing adequacy of coverage.

Ecuador had begun to work with the World Bank to establish a “well-targeted, efficient, and sustainable” social protection system in 2018, “in the context of imminent fiscal adjustments.”[180] The World Bank approved the $350 million Social Safety Net Project in April 2019, shortly after the IMF approved a $4.2 billion loan conditioned on austerity.[181] Previous efforts to more narrowly target Ecuador’s safety net had resulted in dramatically reduced coverage, eroding efforts to reduce poverty, according to the World Bank.[182] This cash transfer program was intended to increase coverage, while improving targeting, and aimed to reduce extreme poverty from 8.7 percent in 2018 to 3.5 percent in 2021. The IMF program approved in 2020 to replace the 2019 program incorporated this cash transfer program into its conditionalities, setting a target of covering 80 percent of the poorest 30 percent of Ecuadorians, which roughly coincided with the rate of poverty prior to the pandemic.

However, as in many countries, the pandemic and other economic shocks significantly increased poverty in Ecuador. By the time the IMF conducted its first review of the program, published in December 2020, the pandemic had pushed around 1.5 million Ecuadorians into poverty, raising the poverty line from 30 to 38 percent of the population.[183] This means that even if the IMF coverage targets were achieved, which would not happen for another two years, the expanded social safety net would cover roughly 4.2 million people, while leaving out more than one-third of households living under the poverty line, some 2.5 million people. Moreover, one way in which authorities were expected to achieve the targeted expansion is by removing ineligible families from the social assistance registry; to reduce “leakage,” in IMF parlance. But this would include removing benefits from those families whose income placed them above the 30^th percentile but below the poverty line; some 1.5 million people. The IMF review flags this as a concern and urges the authorities to avoid such an outcome, although there are no specific measures to ensure they do.[184]

Poverty rates have since decreased in Ecuador, returning to roughly the same level as prior to the pandemic. While it is positive that the increase in poverty abated, they remain high. Most strikingly, around 70 percent of people in rural areas were considered multidimensionally poor, a measure that takes into account numerous factors such as access to health services, education, and clean water, in both December 2019 and 2022.[185] Speaking to Human Rights Watch, Dr. Andrés Mideros Mora, professor of economics at Pontificia Universidad Católica del Ecuador, explained that the failure to achieve any reduction in poverty from the 2019 baseline demonstrates the limitations of these types of programs for achieving deep structural changes needed to meaningfully reduce poverty.[186] He also cautioned that the transfers are very low and questioned whether they are sufficient to compensate for gaps in other forms of support or public services, which risk being further eroded by austerity cuts.[187] According to the World Bank, Ecuador’s spending on education decreased from 4.6 percent of GDP in 2018 to 3.7 percent in 2021.[188]

Moreover, the sharp spike in poverty during the first two years of the pandemic illustrates a key problem with setting coverage of social safety nets to the poverty line: it treats poverty rates as fixed when in reality economic shocks and booms mean that these numbers swell and wane and, at a minimum, governments need the flexibility to expand coverage as necessary. This is true both of poverty rates and individuals’ or households’ economic status. The graph below prepared by Development Pathways demonstrates the changes in socioeconomic status over just two years in Vietnam and Indonesia.

Poverty Lines Are Not a Good Proxy for Need

Even if targeted programs were able to adapt to poverty rates, the focus on only covering people in poverty is too narrow to ensure that all people are able to realize their economic and social rights. National poverty lines are often arbitrary or set well below what is required to protect people’s rights.[189] For example, Ecuador’s national poverty line is based on the cost for a specified basket of goods adjusted annually for inflation. In December 2022, the poverty line was set at $88.72 per capita per month, which amounts to $344 monthly for the average family of 3.88 people.[190] According to the calculations of the Global Living Wage Coalition, a research and advocacy coalition that advocates for living wages, a family in rural southern coastal Ecuador, where the cost of living is relatively low, would need to earn at least $482 per month (or $5,784 per year) to meet its needs. Even this measure is limited, as averages can’t capture the range of people’s particular circumstances and the gaps the prevent them from the full realization of their rights. But even by this measure, it means many people are above the poverty line, but do not earn enough to meet their needs.[191] In 2020, the median income was $5,444, according to the IMF.[192] Put another way, while around 25 percent of Ecuadorians lived in households with income below the poverty line in December 2022, 38 percent were considered multidimensionally poor, a measure that tracks deprivation in multiple areas and therefore better captures people’s circumstances than the poverty line.[193]

In addition, in countries where poverty lines are not accurately adjusted for inflation, they can fail to capture a sharp decrease in purchasing power that can undermine people’s ability to realize their rights. This limitation is especially salient in the current context where prices, particularly for fuel and food, have dramatically increased in many countries. In Ecuador, while the poverty line is adjusted for inflation, according to Dr. Mora, because the adjustment is based on price increases for a specific basket of goods in nine cities, it fails to account for higher prices in rural areas, which have significantly higher rates of poverty.[194]

The limitations of using poverty lines as a measure for need—not to mention for rights—to determine social protection coverage add to myriad other problems that research on targeted programs has surfaced, many of which the IMF technical note on social safety nets acknowledge. Targeted programs routinely have extremely high rates of exclusion errors (that is, people who are eligible but do not receive support) and inclusion errors (that is, people who are ineligible receive support). They also have high administrative costs that can reduce funds available for benefits. For example, administrative costs account for 17 percent of combined spending on social safety nets among African countries.[195] Studies have also shown that targeting social protection only to people in poverty leaves out large segments of the population who are near-poor and vulnerable to poverty, a phenomenon known as the “missing middle.”[196] In doing so, targeted programs can not only exclude people who need support to realize their rights, but also fuel social resentment that erodes public and political support for the programs.[197]

An overly narrow focus on people living under national poverty lines also leads the IMF to underestimate the program’s harmful impacts in its poverty and inequality analysis.[198] Because the cash transfer expansion, in theory, would raise the effective incomes of the poorest 30 percent, the assessment concludes that “taken together with the expansion of social assistance, the envisaged tax package would help reduce income inequality.”[199] However, this papers over the losses experienced by people near poverty or in the middle class. Indeed, as noted, the staff assessment posits that reducing the public wage bill would decrease inequality because many public-sector salaries are above the median. The intent of addressing income inequality should not equate to lowering the wages of people earning slightly above the median wage. On the contrary: cuts and caps on public wages can erode the quality or availability of public programs essential to rights. Similarly, the assessment finds that the VAT hike and expansion of the personal income tax reduces inequality, even though it also hurts people who are in or near poverty but are not covered by social protection programs.

Sequencing and Enforcement

In addition to raising questions about the adequacy of social spending floors, the gap between paper and reality in the expansion of Ecuador’s social protection program brings to light problems of sequencing and enforcement. The impact of adjustments is often immediate, with prices increasing as soon as subsidies are removed or value-added taxes implemented. But building and expanding social safety nets takes time, particularly when they are targeted programs that depend on complex bureaucracies. In the case of Ecuador, one year into the program, the government had only extended coverage to 30 percent of the poorest 10 percent of Ecuadorians, meaning that well over 1 million of the poorest Ecuadorians had yet to receive any support, even though the analysis projecting the impact on poverty had assumed they would. The IMF staff report explained the delay as “partly reflecting challenges in reaching the most vulnerable households in some peripheral areas and urban pockets of poverty.”[200] Because the expansion was intended to be gradual, despite this gap the government met—indeed exceeded—the coverage targets it was required to achieve by the first review, which was published in October 2021. To afford the government more time to achieve the final goal, it delayed completion by a year to December 2022 and reduced required coverage to 70 percent of the bottom 30 percent, with at least 65 percent coverage among the poorest 10 percent. By the time the final review was published in December 2022, the government was still unable to achieve targets in three districts, although it had achieved the overall target of covering 80 percent of the bottom three deciles.[201]

The difficulty in achieving even modest targets, and the time it takes to achieve them, reflects a problem inherent in the IMF’s model of compensating for immediate harms of adjustments with expansions in targeted social protection programs. Unless the reforms are carefully sequenced to take place after adequate expansion has taken place, there will be a gap where people on low incomes will be left without the means to absorb higher costs associated with many IMF-driven reforms. Moreover, because the social spending floor is incorporated as indicative targets rather than performance criteria, as they are in nearly all programs, noncompliance does not require a special waiver from the IMF Board. An analysis by Oxfam of 17 IMF loans approved in 2020 and 2021 found that 22 out of 63 social spending floors were not met.[202]

Funding Social Protection Improvements

Few people argue that less spending on social protection is better; rather, they point to mounting government debt and limited revenues to justify strictly means-testing programs. For example, the IMF technical note on social safety nets (SSN) instructs staff to make communication central to any plan to reform social safety nets due to fiscal pressures:

Where SSN reform is motivated by the need to address fiscal pressures, communication of reforms needs to be explained in terms of trade-offs within the constraints of government’s financing capacity. For example, in the context of the need for fiscal consolidation, communication could emphasize that reforms are focused on reducing leakage of benefits to higher-income households while also addressing any existing gaps for protecting more vulnerable households.

This approach ignores the important spillover effects of social protection; both what is gained by increased spending and what is lost by reducing it. A recent study of child grants in Lesotho found that each loti (LSL) transferred to a eligible households increases total nominal (cash) income in the local economy by 3.11 LSL and real (inflation-adjusted) income by 1.67 LSL.[203] It also found that even households that did not receive the cash transfers saw benefits due to higher spending.[204] By the same token, as noted in its internal post mortem on its response to the 2008 Financial Recession, the IMF found that cutting government spending had led to economic contraction in advanced economies instead of the growth the IMF had anticipated due to forecasters underestimating the “fiscal multiplier effect;” that is, the ripple effect of increasing or reducing government spending on every sector of the economy.[205]

Moreover, in the context where nearly every IMF program relies on fiscal consolidation, the view that it justifies reducing or strictly limiting spending on social protection, or that the harmful impacts of fiscal consolidation can be adequately mitigated through targeted cash transfer programs, can limit IMF staff from considering alternative pathways to economic recoveries that enhance, rather than erode, rights. Alternatives are available. Ortiz and Cummins lay out nine ways of enhancing fiscal space to avoid austerity.[206] These include strengthening progressive revenue generation and tackling illicit financial flows; objectives which have particular resonance following a pandemic that saw inequality soar, with a small group of well-placed individuals and businesses experiencing a windfall while hundreds of millions of people were pushed into poverty.[207] Yet many governments have reduced personal or corporate income taxes in recent years.[208] Sri Lanka, for example, implemented steep tax cuts in 2019 despite already having one of the lowest tax-to-GDP ratios in the world, a move that is widely regarding as contributing to the economic crisis roiling the country today.[209]

Even within the context of fiscal consolidation, the overarching emphasis on reducing or containing social protection coverage through targeting comes at the expense of an alternative hinted at in the technical note but never fully explored: Instead of investing in the administrative architecture to target social protection coverage, governments can provide broad coverage and focus their efforts on improving progressive income tax structures to “claw back” benefits from high-income households.[210] This approach avoids the risk of exclusion errors and other problems associated with targeted programs and directs precious public dollars toward developing a reliable progressive tax system. In practice, IMF programs rely significantly on value-added taxes to raise revenues; 19 of 33 programs reviewed included advice or conditionalities on raising VATs or eliminating exemptions such as on food.

An approach that looks at equitable revenue generation should go beyond official tax rates: companies have long exploited global tax regimes to shift profits to reduce or eliminate their tax bills, depriving the countries where they operate of an important source of revenue. And many wealthy individuals and companies avoid paying even those taxes to which they are legally subject through legal or illegal means. Tax havens collectively cost governments between $500 billion and $600 billion a year in lost corporate tax revenue, and they are a particular drain on low-income countries.[211] Estimates of individual wealth stashed in tax havens is even greater: $8.7 trillion to $36 trillion depending on the estimate.[212] Reducing corruption, such as by tackling abuse in public procurement, can also make a significant difference, with one study finding it could raise $1 trillion in new public revenues, particularly for low-income countries.[213] In contrast, an ILO paper calculated that the global cost of establishing universal social protection in 2019 would be US$792.6 billion, including administrative costs.[214]

 

III. Experiences from the Ground: The Case of Jordan

The Fund’s approach of continuing to pursue its traditional line of structural adjustments with governments while mitigating their impact on people on low incomes through targeted social protection is central to its response to the pandemic, but it did not begin with it. This approach has been a hallmark of several major IMF programs over the past decade, including in Pakistan, Egypt, and Jordan. Having the benefit of time, these programs offer a valuable window into whether doubling down on this approach in the wake of the pandemic is consistent with the IMF’s commitment to building more inclusive economies with strong social protection programs.

This chapter documents the real-world impact of a decade-long series of IMF programs in Jordan. Based on a close analysis of relevant documents as well as around 50 interviews with people on low incomes affected by the program and 20 relevant experts, it offers an illustrative case study of how people experience the consequences of traditional adjustments paired with new allocations for targeted social spending protected by a spending floor.

IMF and Jordan

Samer, standing at the entrance of his drinking water shop in Marka, a low-income neighborhood in the Jordanian capital of Amman, recounts the changes to his business since opening in 2009.[215] First, he says, fuel subsidies were cancelled in 2012. The wholesale cost of water, which is delivered in a tanker truck, went up, as did the cost of his deliveries, but he did not receive any government support to offset these increases. “Now we need [to deliver] at least 10 jugs to justify the cost of the fuel.”[216] He didn’t raise his price for a 10-liter jug, he said, but he had fewer customers as people struggled to afford rising prices generally.

Then, in 2018, bread subsidies were removed and new sales taxes introduced, which cut further into his sales as people struggled to make ends meet. At the same time, his electricity costs started to increase, further cutting into his profits. “Every year is worse than the year before,” he told Human Rights Watch, but the Covid-19 pandemic dealt him yet another blow, reducing sales further by roughly one half. “People have no money,” he said. “When I started in 2009, I sold 120 JD per day (US$170) [worth of water], now it’s 25-30 JD ($35-42). The water tanker used to come every three days, now it’s every nine days.”

As he’s speaking, a customer fills a 10-liter jug and hands him money. He explains that the customer took a jug a few days earlier but didn’t have the money to pay so he’s now paying for both jugs. He tells Human Rights Watch that he has a rule not to refuse people who ask to buy on credit “because you can’t tell a person they can’t have water.”

Jordan’s economy has faced many challenges over the past decade, but the timeline Samer described, up to the point of the pandemic, tracks with a comprehensive economic reform program the government enacted and financed under a series of IMF programs, worth a combined $4.4 billion, that began in 2012. At the time, Jordan’s debt had climbed to 80 percent of GDP, largely due to an increase in energy costs and the lingering effects of the 2008 Financial Crisis.[217] The effects of regional instability also had complicated economic implications for the country, hurting its tourism sector and bringing hundreds of thousands of Syrian refugees into the country. The programs, which followed a series of IMF programs in Jordan spanning from 1989 to 2004, were designed to reduce Jordan’s debt,[218] including by removing or reducing subsidies on fuel, electricity, and food, and by increasing sales taxes.[219] Despite Jordan implementing these reforms, its debt continued to climb, reaching more than 97 percent of GDP in 2019, before it sharply rose again due to the Covid-19 pandemic, topping 110 percent by the end of 2022.[220]

But even if the IMF-driven measures had achieved their objective, both the IMF and Jordanian authorities acknowledged the pressure they would place on low-income Jordanians, and each of the loan programs included language about the importance of mitigating these impacts through improvements in social protection, as described below. Yet the only significant effort to do so came in 2019, when the government, with the support of a combined $2 billion World Bank project, established a targeted cash transfer program called Takaful.[221] This chapter will examine how effective Takaful has been at achieving this goal since it is a paradigmatic example of the IMF’s efforts to simultaneously embrace fiscal consolidation and social protection.

2012-2018: Higher Prices, Scant Support

Until the establishment of Takaful, The National Aid Fund (NAF), established in 1986 to operate as an administratively and financially independent entity, oversaw the country’s main social safety net, a program that provides small cash transfer payments to certain groups of people, such as orphans and older people.[222] In 2018, the year before Takaful was established, it reached around 330,000 people (in 100,000 households) even as the poverty rate among Jordan’s population of 11 million was 15.7 percent that year; meaning, it benefitted less than one in five people living in poverty.[223] In 2011, Jordan established unemployment insurance, although it is limited to formal sector employees who have paid contributions to social security and the benefit amount is based on these contributions.[224] As of August 2019 it covered 630,000 people, less than one-quarter of the total labor force.[225]

In December 2012, the IMF approved a $2 billion, 36-month Stand-By Arrangement that imposed the removal of fuel subsidies among other measures that were intended to drastically reduce government spending. These measures, which came at a time when public anger over the government’s handling of the economy and perceived corruption was already at a boiling point, were met by widespread protests.[226] The loan agreement states that authorities “noted the need to strengthen the social safety net to shield the poor from the impact of price increases.”[227] However, it did not include a social spending floor or any specific measures to achieve this goal, nor did it calculate the anticipated impact on people on low incomes. An IMF review of the program the following year noted that retail prices rose 14 percent for gasoline, 54 percent for liquid petroleum gas, and 33 percent for kerosene, which in turn raised prices more broadly.[228] Separate IMF research from 2015 that looked at the impact of fuel subsidy removal in a number of countries found that household costs rose 5.8 percent following the removal of fuel subsidies in Jordan.[229]

In 2016, the IMF approved a second loan, a $723 million, three-year Extended Fund Facility that was similarly conditioned on fiscal consolidation. To achieve the fiscal targets, the government removed bread subsidies in 2018, causing the price of bread to rise between 60 and 100 percent, an increase of around 0.16 JD ($0.23) per kilogram.[230] The government also implemented broad tax reform that relied heavily on increasing sales taxes and reducing the threshold of income subject to income taxes. Sales taxes were increased to between 10 and 20 percent on essential food items including dairy products, eggs, and fruit, that had previously been taxed at between 0 and 8 percent.[231] A new income tax law reduced the threshold for taxable income from 12,000 JD ($16,916) to 9,000 ($12,687), and progressively increased the rate on the next 20,000 JD ($28,193) annually to reach 25 percent, compared to 20 percent previously, with income over 1 million JD newly taxed at 30 percent.[232] Ahmad Awad, the head of Phenix Center, a Jordanian NGO focused on economic issues, described it as a “progressive tax for the middle class but a flat tax for the wealthy.”[233] Awad also noted that the income tax reform maintained the government’s reliance on sales tax, which made up roughly three-quarters of all taxes in 2022.[234] Inflation increased by nearly 5 percent that year and the reforms sparked protests across the country that ultimately led to the resignation of the prime minister.[235]

As with the previous loan program, this one stressed that “successful implementation of the program will require cushioning the potential impact of fiscal consolidation and structural reform on the vulnerable segments of the population.”[236] This time the program included “an indicative floor on social spending to shield low-income groups from the effects of fiscal adjustment.”[237] The floor was defined as a category of social spending already tracked by the central government budget that includes transfers for sickness and disability; older age; family and children; housing; and other social protection related costs.[238] However, the Jordanian government did not meet the social spending floor in either of the program’s two published reviews, in which IMF staff assess program compliance before releasing the next tranche of funds.[239]

Since the floor was an indicative target rather than a performance criterion, this failure did not require a Board-approved waiver for the program to continue, leaving little incentive for the Jordanian government to ensure compliance. In 2016, the floor anticipated that the social protection budget would increase by 230 million JD over the previous year, yet the first review found that the government’s actual spending only increased by around 60 million JD.[240] The second review, published in 2019, similarly found that spending in 2017 and 2018 consistently fell short of the indicative targets of the social spending floor; by 89 million JD in 2017 and by 55 million JD by June 2018.[241]

It is also unclear where any increase in social spending under the floor went or any real-world impact. The NAF’s budget increased only slightly between 2012 and 2018. In fact, between 2012 and 2017 the budget allocated to targeted social assistance decreased by 9 percent, or JD 21 million, even as non-targeted subsidies decreased by 88.4 percent, or JD 788 million.[242] A separate analysis in a paper by the Friedrich Ebert Stiftung found that public spending on health, education, and social protection stayed roughly the same as a percentage of the overall budget during this period, suggesting that the savings from subsidies was not directed to health or education either.[243]

With such a threadbare social protection system, the public was left largely unprotected to weather not only a sputtering economy that led to stubbornly high unemployment, but also a series of painful IMF-driven economic reforms that drove up prices.[244] Human Rights Watch has previously documented how this contributed to an explosion of personal debt in Jordan, which, when people are unable to repay, can lead to imprisonment in certain cases because of Jordan’s draconian laws that criminalize overdue promissory notes and bounced checks.[245]

2018-2021: Takaful, Covid-19 Relief, and Another IMF Program

In February 2018, protests broke out to oppose the increase in prices due to the removal of bread subsidies and increase in sales taxes, echoing demonstrations that roiled the country in 1996, when the government had previously lifted bread subsidies, only to restore them soon after due to public pressure.[246] This time, the government increased NAF’s funding by 10 million JD and introduced a one-time subsidy ranging from 27 to 33 JD for Jordanians and some non-citizen Jordanian residents with annual incomes below 6,000 JD.[247]

The same year, the World Bank approved the first part of a $2 billion program that included funding for establishing a targeted cash transfer program explicitly designed to buffer people from the impact of these structural adjustments. According to the program documents, “the adverse effects of potentially higher prices will be mitigated by an improvement and expansion of the National Assistance Fund.”^[248] Soon after, the government released a National Social Protection Strategy for 2019-2025, with funding and support from UNICEF, that introduced Takaful, the country’s first cash transfer program designed for the working poor.[249]

The strategy acknowledged that, until then, the government had largely relied on subsidies, such as on fuel, water, and wheat, as a social safety net, as well as small cash transfer programs meant for those unable to work. Takaful marked an important step in “shifting social assistance resources away from inefficient subsidies towards poverty targeted programs.”[250] Introducing or relying on existing highly targeted cash transfer programs is emerging as a hallmark approach for the Fund, working in partnership with the World Bank, to reconcile its recent emphasis on building inclusive economies with its traditional fiscal adjustment policies, including removing or reducing subsidies. Indeed, Jordan’s current Minister of Finance, Mohamad Al-Ississ, criticized previous IMF programs’ heavy reliance on spending cuts and regressive taxation, saying “they did not live up to the hopes that both our people and the Fund aspired to because there was a missing element of ownership and trust.”[251] While Al-Ississ praised the 2020 IMF program he helped negotiate and the increased attention to social safety nets, his comments reflect the extent to which Fund advice is often interwoven in government policies.[252]

In Egypt, for example, the government established two new cash transfer programs—one, also called Takaful, targeting poor families with children and another, called Karama, targeting older people and people with disabilities—in 2015. As in Jordan, it was supported by the World Bank and intended to mitigate the impacts of IMF-driven reforms, “which included removal of energy subsidies, the adoption of a flexible exchange rate and the introduction of a new VAT tax.” [253] According to the World Bank, Takaful and Karama “aimed to mitigate the potential short-term, negative impacts of the reforms on the most poor and vulnerable segment of Egypt’s population.”[254]

This strategy is also reflected in a new IMF loan program in Jordan approved in August 2020, months after the Covid-19 pandemic enormously exacerbated the challenges that low-income Jordanians were facing. The agreement, which is discussed in detail below, called for fiscal consolidation while making it “an important objective” to counter the impacts of the reforms with “adequate protection of the poor and vulnerable,” and cited “cooperation with the World Bank and UNICEF” to enhance the social safety net as a means of achieving that goal.[255] An IMF review of the program conducted in 2022, which mandated the removal of fuel subsidies after the government restored them to cope with a spike in prices due to Russia’s 2022 invasion of Ukraine, makes the link more explicitly. For example, the Executive Board Assessment of the program notes that “Directors stressed the need to replace costly fuel subsidies with targeted support to protect vulnerable households within the approved 2022 budget envelope.”[256]

A key change in the 2020 program is that the social spending floor, which was defined to be broader than the floor in the previous program and includes new spending on the cash transfer program, was made into Performance Criteria that require a Board waiver if unmet.

At the same time, the effectiveness of the floor depends on the amount where it is set. As discussed below, some reviews of the Jordan program lower the floor from where it was the previous year, indicating how they can also be a tool to exert pressure to decrease social spending by unwinding government support even as the economic impacts of the pandemic, inflation, and other economic stresses, including IMF-driven policy changes, are still acutely felt. And, as the Board Assessment above hints at, the targeted cash transfer programs that are meant to mitigate these impacts can act more as a constraint on social spending, allowing enough to cover only the bare minimum and leaving too many people exposed.

Takaful

Takaful is an automated cash transfer program intended to provide income assistance to the poorest Jordanians. Its original budget was $200 million over three years funded by the World Bank, and UNICEF has been closely involved in the program’s design and implementation. All families under the poverty line are eligible but it is funded to cover far less than the number of eligible families. In 2022, it covered 120,000 families, which amounts to less than one-fifth of the roughly 24 percent of Jordanian families who now live under the poverty line.[257]

To apply to the program, applicants fill out an application questionnaire, which is available online or can be filled out at one of NAF’s offices or temporary registration centers and elicits information about household demographics, spending, income, and assets. Applicants are first screened to determine if they meet basic eligibility criteria: for example, whether they are a Jordanian citizen with a national ID number, if they are receiving other NAF benefits, and whether they are under the national poverty line. Applicants identified as eligible do not necessarily become beneficiaries, since the scope of the program is set based on available funding. Instead, an algorithm ranks eligible applicants from poorest to less poor based on 57 socio-economic indicators, using a combination of data drawn from numerous agencies as well as information reported on the application questionnaire. Examples of indicators or sub-indicators that are weighed to calculate an applicant’s ranking according to NAF officials are: family size, illness, governorate, number of children who are working age, the value and age of cars owned by the family, size of the house, electricity usage, and expenditures.[258] Applicant families ranked poorest are selected as funding allows.[259] In the beginning of the program, the algorithm calculated the amount awarded to each household based on a complicated equation, but it has since been simplified so that households receive between 40 and 136 JD per month depending on the number of people.[260] Another change to the program is that households are “refiltered” each month to ensure continued eligibility. Under the new system, benefits are supposed to arrive monthly.

Takaful’s timing came at an unexpectedly critical moment for Jordan. The program launched in May 2019 and the first round of payments went to 21,084 families in February 2020, just as the novel coronavirus was beginning to bring economic activity to a standstill.[261] With the help of additional emergency loans from the IMF and World Bank, Jordan was able to quickly expand Takaful to support informal workers impacted by Covid-19.[262] Nearly half of workers in Jordan are informal and most are not covered by the country’s unemployment insurance scheme.[263] In the spring of 2020, the government launched Takaful-2, a one-year program that sent assistance to 240,000 day laborers. Households with less than three members received 72 JD per month, paid every one to two months, and larger households received 132 JD. The program was extended at a smaller size—covering 160,000 workers—through December 2021, after which it expired despite the lingering impacts of the pandemic.

To make matters worse, Russia’s invasion of Ukraine in February 2022 compounded the economic hardship people faced by raising the cost of fuel and grain. To help mitigate these impacts, the government expanded Takaful to reach a total of 120,000 families. It also restored some fuel and bread subsidies, but under the ongoing IMF program the fuel subsidies were again removed and the government committed to work with the Fund’s technical assistance to explore “better targeting” the food subsidies.[264]

2020 IMF Loan Program

The $1.3 billion, three-year program, which is currently in its final year, reflects the countervailing impulses to increase social spending—and social protection in particular—and to decrease debt by cutting government spending.

The program continues fiscal consolidation totaling 4 percent of GDP over the program period, following a brief pause to allow for crisis-related spending. It builds on previous IMF programs in several respects and introduces some new reforms. Like many IMF programs approved since the pandemic, it calls for freezing public sector hiring except in health and education. While women are always more likely to be impacted by such freezes, this is likely especially true in Jordan: education, health care, and public administration represent 30 percent of total employment but hire 68 percent of all employed Jordanian women, according to a study produced by Harvard’s Center for International Development.[265] While health and education are excluded from the hiring freezes, as the study points out, “given fiscal restrictions, it’s unlikely that these sectors can grow enough to absorb the large increase expected in qualified female labor in the coming years."[266]

At the same time, IMF programs in Jordan, including the current one, identify Jordan’s low rates of female labor participation as an obstacle to economic growth and include a number of measures to improve their participation, including by improving and implementing regulations requiring firms to make daycare available to employees.[267]

The 2020 loan program also includes two reforms that are intended to reduce the burden on businesses in Jordan. It expands a measure enacted under the previous program that allows certain employers to reduce their social security contributions for employees under the age of 28 to include all start-up companies, which can erode the long-term sustainability of the social security system. The program also reforms Jordan’s cross-subsidization of electricity to reduce tariffs for businesses by raising them on certain households. A new tariff structure, discussed in more detail below, was introduced on April 1, 2022.[268] The program also signals possible water tariff reform, as the government is required to “study the water tariff structure both for corporate and households and explore options for a restructuring of water tariffs and their potential social impacts.”

In 2022, fuel and food subsidies once again came into focus, after Russia’s war against Ukraine significantly raised prices on both. To help people cope, the government restored some of the subsidies it had removed under previous programs. In line with requirements laid out in the fourth staff review of the program, the government began to remove the fuel subsidies in May 2022, and they were fully phased out by the end of the year.[269] Some food subsidies remain in place, but the government committed to work with IMF technical assistance to, among other things, “help improve targeting.”[270]

In addition to reducing expenditures, the program also emphasizes the need to improve revenue generation, and includes several structural benchmarks related to closing tax loopholes and tackling tax evasion.

At the same time, the program pays greater attention than the previous one to social spending. It includes a social spending floor, this time as Performance Criteria, “to ensure that any necessary fiscal consolidation to preserve debt sustainability does not burden the poor, but rather promotes equity, and protects Jordan’s future growth potential.”[271] The floor is defined to include non-wage health and education current expenditures, the NAF social protection program (which includes Takaful), and the school feeding program. While more precise than the previous definition, it does not track existing budget lines, and the program does not calculate spending in previous years on the categories included in the definition, leaving the public unable to track spending relative to previous years.

Moreover, even if the floor does represent an increase over previous years, that alone does not speak to its adequacy given that, as noted by a June 2021 IMF review, “the budget allocation for social assistance has been historically low compared to regional and global peers.” [272] Despite this low baseline spending, as well as the backdrop of a pandemic, inflation, and additional economic reforms that further raised the cost of living in ways that can harm the rights of people in poverty, the program repeatedly expects immediate and drastic cuts to the increase in spending in response to the pandemic. Reviews of the program raise the floor in light of these challenges, but the initial targets could indicate Fund impatience to drastically lower social spending.

When the program was first approved in August 2020, a time when much of the world was still in the grips of the initial waves of the Covid-19 pandemic, the social spending floor for 2020 was set at 514 million JD ($725 million).[273] It was revised upward to 804 million JD during the first review, which was published in January 2021, to reflect higher spending as part of the government’s Covid-19 response; actual spending was even higher at 883 million, of which 240 million JD went to cash transfers.[274] In 2021, the floor was set for even higher than that—943 million JD—to allow for an increase in health spending; actual spending was 1,028 million JD, although spending on cash transfers decreased to 184 million.[275] However, a review of the program published in July 2022 lowered the floor to 793 million JD, reflecting Fund eagerness for governments to begin to unwind Covid-19 spending.[276] As with previous reviews, the floor was revised upward to 896 million JD.[277] This represents a decrease of over 12 percent from actual spending the previous year despite inflation reaching over 5 percent in 2022.

In millions of JD

The fifth review, published in January 2023, even more aggressively attempted to bring down social spending by lowering the floor to 532 million JD, which would represent a 40 percent decrease from the previous year’s floor.[278] The sixth review, published in June 2023, made clear that this attempt was unsuccessful, and actual spending was higher in 2022 than the previous year, although actual spending on cash transfers was lower than projected.[279] A new floor was set; far higher than the initial floor for 2023, but around 20 percent lower than actual spending the previous year. In the years ahead, spending on cash transfers are projected to increase slightly to 280 million JD by 2028.[280]

The steep drop in the social spending floor—essentially returning it to pre-Covid-19 levels—and projected stagnant spending on cash transfers is at odds with the program’s recognition of how Jordan’s social spending in general, and Takaful in particular, fall short. As already noted, the first review called attention to Jordan’s low rates of social spending relative to peer countries. Later program reviews explicitly link the removal of fuel subsidies with investing in a strong safety net, while acknowledging that Takaful is inadequate. For example, the fourth review calls Jordan’s safety net “very efficient in reaching the most vulnerable citizens” but acknowledges that “the budget allocation for social assistance has been historically low compared to regional and global peers, raising the question of adequacy of the social safety net.”[281] It goes on:

While post-COVID, budgeted social spending is above previous levels, and spending in categories monitored under the EFF program floor on social spending has been increasing, there is scope for further improvement. As a first step, given the persistently high unemployment, and high demand (with over 400,000 households applying in the past six months), the newly launched unified cash transfer program could be expanded well above the current size of 120,000 families.[282]

In a similar vein, as part of the fifth review of the program, the government acknowledges “unmet needs for social protection”:

We increased the number of families covered through the unified cash assistance program reaching 120,000 families in 2022, although there are unmet needs for social protection, consistent with the high rates of poverty and unemployment.[283]

In this light, it is clear that dropping the social spending floor by 40 percent and projecting relatively stagnant levels of cash transfers fails to address the need for an increase in spending to protect people’s rights from a relentless cascade of economic woes.

“I’m Dying of Hunger and I’m Not Getting Aid”

The story of Takaful is a testament both to the powerful impact the IMF can have when it builds government social protection into its programs and the problems that ensue when its vision of what is needed is overly narrow. After having largely overlooked social protection in the first seven years of its IMF-driven economic and fiscal reform program, Jordan’s eventual investment in building the infrastructure for a reliable cash transfer program enabled it to quickly disburse funds once the pandemic hit. Many governments that did not have such infrastructure in place were left unable to disburse funds at all or were forced to rely on leaky systems rife with patronage and political interference.[284] At the same time, as recent IMF staff reports appear to acknowledge, Takaful is still a very limited program that has not done nearly enough to buffer people from the cost of the IMF’s program adjustments.

The national poverty line in Jordan is set at 68 JD per month per household member and the average household is around five people, although poorer families tend to be larger.[285] Importantly, this amount, which was set in 2010, is calculated based on expenditure rather than income, so worsening of deprivation due to an increase in the cost of living may not be fully reflected in the official poverty rate.[286] The minimum wage is 260 JD per month, which is 80 JD short of the poverty line for a family of five with a single breadwinner. Such situations are common as only 13.2 percent of Jordanian women work, one of the world’s lowest rates of female labor participation.[287] To put this amount in perspective, Abdelhamad, an employee at a coffee shop whose income had just been raised to 250 JD, told Human Rights Watch that “it’s impossible for our expenses to be any less than 350 JDs.” He described his expenses for his family of two children: “My rent is 140 JD, my phone, electricity, water bills and other expenses are 320 JD. Each month, we just trust in Allah and hope we can cover our expenses. Some months my siblings abroad help out.”[288] Yet his expenditures are well above the poverty line.

Of Jordan’s roughly 11 million citizens, 15.7 percent lived under the poverty line in 2018, although some experts believe the real number even then was much higher.[289] A change in how Jordan calculates poverty in its Household Income and Expenditure Survey between 2011 and 2018 means it is not possible to compare poverty rates over this period, but an independent analysis found an increase of 8 percentage points.[290] Since the pandemic, the official number has increased to around 24 percent.[291] The 2015 Census found that 600,000 children in Jordan are multidimensionally poor.[292] These numbers exclude the large number of noncitizens living in Jordan, including 673,957 registered Syrian refugees, around 500,000 Palestinians without Jordanian citizenship, and nearly one million migrant workers.[293] These groups are disproportionately poorer and are equally impacted by price increases yet authorities actively exclude them from accessing Jordan’s safety net, instead requiring that they rely on chronically underfunded international humanitarian agencies to make up the difference.[294] For example, the Office of the United Nations High Commissioner for Refugees’ (UNHCR) $405.7 million budget for Jordan was only 59 percent funded in 2021,[295] and the United Nations Relief and Works Agency’s (UNRWA) $38 million appeal for Jordan in 2021 was only 53 percent funded.[296] For Syrians and Palestinians, support is available through UN agencies, although the funding tends to be irregular and inadequate, subject to donor commitments. For migrant workers, there are no equivalent support systems, and they are excluded from accessing Jordan’s safety net. Jordanian women married to non-citizens or who have non-citizen children are also excluded from some forms of support, and their husbands and children are entirely excluded.[297]

The economic hardship people face has multiple causes, including extremely low wages and limited unemployment or disability benefits. Jordan’s minimum wage is based on a monthly salary, effectively exempting hourly workers. Informal workers and day laborers—who make up almost 60 percent of the total workforce—earn particularly low wages and are not protected by minimum wage standards.[298]

Everyone Human Rights Watch spoke with described serious struggles due to rising prices, including those who were able to benefit from Takaful. Many of these interviews were conducted in October 2021, a time when the economic shock of the pandemic was acutely felt but before global inflation, in part due to Russia’s 2022 invasion of Ukraine, greatly exacerbated people’s financial challenges. Many had applied to Takaful but did not receive any benefits despite dire need; most who had received support benefited from the Covid-19 expansion which has since expired. People interviewed emphasized that the sharp rise in prices began before the pandemic, although, as in many countries, it has gotten much worse due to pandemic-related supply chain issues. Several people specifically cited the removal of subsidies in 2018 and 2019 as the moment when their economic hardship worsened, and a few specifically blamed international institutions or the IMF for these reforms.

Most often, people singled out two issues when asked about the impact on their lives: food and electricity. Many people described cutting down on the amount and type of food they and their families eat, forgoing or drastically reducing animal protein because they could no longer afford their usual quantities of meat, chicken, and fish. Many also described being unable to afford their electricity bills, with many facing persistent power cut-offs due to unpaid bills.

Moreover, everyone interviewed expressed confusion about the available aid programs. No one was able to explain eligibility requirements, and many noted that they didn’t understand why some people received benefits and others didn’t. Several described receiving an error message from the online registration portal but not understanding the reason for it or being denied support despite living in dire poverty. In some cases, they believed their chances of receiving support were diminished because of an asset that did not actually provide them with a means to support themselves; an old car or a fallow lot of inherited land in the countryside, for example. The lack of clarity led some who were denied benefits to express resentment at those who did get support. People who did receive benefits were generally unsure how the amounts were calculated, or when, how often, or for how long they would be disbursed. Several described receiving only one or several irregular payments that do not appear to align with the rules of any cash transfer program.

Food

Ibrahim lives with his wife and four children in Zarqa, a town around 30 kilometers northeast of Amman.[299] In 2018, he lost his job working as an assistant to a pharmacist in a hospital where he was paid 800 JD a month. But even before he was laid off, he said he was struggling with price increases, even though he and his wife’s combined salary was more than double the poverty rate for their family of six. His wife has a master’s degree in educational psychology but was never appointed to a job as a teacher, one of many people in Jordan who are university educated but unable to find jobs.[300] Instead, she relies on part-time work and earns roughly 190 JD per month.

“Before 2011-12 it was sufficient. You could cover your basic expenses. After that it wasn’t enough but you could ‘make do.’ Fuel went from 14 JD for 20 liters to 24 JD,” Ibrahim explained. Then the price of everything and especially food started to go up. The new sales tax and increase in bread prices coincided with when he lost his job, devastating the family.

“Bread used to be 8 piasters. Even if I didn’t have anything, I could make a pot of tea and buy 2 or 3 kilos of bread. It was enough for us to pass the last 10 days of the month. But now it’s around 42 piasters and much worse quality, so I can’t keep it for a few days to feed the family. It goes stale after one day.”

The pandemic made the situation much worse, Ibrahim said, but at least he was able to receive benefits through Takaful’s expanded program (Takaful-2). Because it’s meant for the working poor rather than the unemployed, he had to put a fictitious salary into the system. In early 2021 he was selected to receive 160 JD per month. The payments, which amounted to 154 JD after fees were deducted, arrived in his mobile e-wallet on March 29; May 3; July 5; and September 9.[301] The last payment noted that it covered July and August, although it was the same amount as previous payments. He did not know when the payments would end or when he would receive the next one.

And the money was far from enough to feed his family. “There’s no bread. We’re hungry. It’s not enough. It costs 5 JD to get to Orange [to get the money] and if there’s traffic, 8-9 JD. It’s not even enough for a week out of the month. The other day, I found my 12-year-old son crying. I asked him why are you crying, at first he didn’t want to say, but I asked again and he said I’m hungry. I told him it is okay, we can go to the fridge and find something to eat. We went to the fridge but there was nothing left.”

Others Human Rights Watch spoke with were going hungry because they had trouble accessing Takaful-2. A day laborer in Marka, a poor neighborhood in East Amman, said he had applied for support several times since Takaful was launched but was denied benefits until the program was expanded to provide pandemic support. He was eligible to receive 270 JD per month, but he said he only received it for two months in April and May 2021, and then the money suddenly stopped. He said he believes it’s because one of his nine children registered with the social security office, although in a meeting with Human Rights Watch, NAF said that this would not immediately cut beneficiaries off from support, since it permits beneficiary households to stay on the benefit for a year in the event that a household member becomes formally employed.[302] “There’s no food. My wife is escaping to her family. There’s no income at all…. It’s been a month since we’ve tasted chicken. The last time I had meat was during Eid. Now we’re going to start eating kittens,” he said sarcastically to describe his level of desperation. “I’m dying of hunger and I’m not getting aid.”[303]

Because most of the interviews were conducted a year and a half into the Covid-19 pandemic, its immediate impacts on income and prices dominated people’s accounts. But nearly everyone emphasized that while “corona” made things much worse, the deterioration in the economic situation began before then. Mohamad is a tailor who works in al-Balad, the local term for downtown Amman, but he hasn’t had much work in recent years. He said he received four disbursements of 130 JD from the National Aid Fund in 2019 but since then he hasn’t received any support. “Since 2011 it’s been very bad, but corona killed us,” he said. “For example, I used to eat chicken or meat every day, now I get it once or twice a week if I’m lucky.”[304]

A man selling vegetables from a cart in the streets of Marka echoed his account. “I can’t afford meat. It started even before corona. Since corona it has gotten much much worse. I have difficulty paying rent. I received a bread subsidy once, then nothing. It was 135 JD in March 2020.” His wife took out a 500 JD loan to help with the increased cost of living and paid it back with 275 JD interest. Separate Human Rights Watch research found that many Jordanians rely on these loans, which have proliferated in the country in the last decade with considerable support from the World Bank and other financial institutions ostensibly as a means of empowering poor people, and particularly women, to pay for daily expenses.[305] They often charge exorbitant interest rates, and failure to pay for any reason can lead to imprisonment.[306]

Several supermarket clerks and vegetable sellers told Human Rights Watch that they have noticed a precipitous drop in recent years of the amount of food people were able to buy, and people have increasingly been requesting to buy on credit. A supermarket clerk in Jabal al-Hadid, a neighborhood in East Amman, said that he had been working there for ten years and noticed a change. “Everything is changed, even before corona.… People are buying less. People used to buy without looking at prices – now they ask. If they used to spend 20 JD buying their kids chocolate, now they spend 5. There’s an increase in the number of people buying on credit, even before corona. People who don’t repay it’s very clear that it’s because they don’t have the money.” The pandemic has made the situation much worse for people, he said. “If it weren’t for me and some friends who owned supermarkets they would have nothing to eat and drink.”

Another supermarket clerk in Jabal al-Hadid had a similar account. “I’ve been selling for 14 years and prices have doubled multiple times. Ten years ago I would sell 1000-1500 JD per day. Today if I sell 150 JD I’m grateful to God. There aren’t that many people coming to buy. So now my income is the worst it’s ever been, even from before corona, but it became much worse since corona.”

Private charities have also been feeling the change, according to a woman who screens applicants for an Islamic charity in Al-Quaismeh, another neighborhood in East Amman. “Every year, the need for assistance has increased, but since corona it is much higher,” she said. “Some people whose house we visited don’t have food to eat. You would be surprised – people don’t have half a piaster for bread. When we distribute chicken, people walk from the ends of the earth to get some.”

Day laborers in al-Ghor, an agricultural area near the Dead Sea, described how the price increases and the pandemic dealt a devastating double blow to their families. Rahma, 32, has six children and has been working in agriculture for 19 years. She earns 10 JD for between seven and eight hours of work, and her husband, who is also an agricultural worker, earns the same amount. She said she applied twice through the portal for assistance, but didn’t get anything, even during the pandemic. “Even charities don’t give us anything,” Rahma said. “We’re basically starving. There was a period of time when we only had two bags of rice and we’d ration it. We spent four to five days during Covid without bread. It costs 5 JD for a chicken for the week.”[307]

Rabia, 31, has two children and earns 5 JD per five-hour shift. She said she has had to cut back on groceries because of price increases. She applied for government assistance and now receives 160 JD every three months, she said, but it’s “not enough at all…. We don’t buy anything, not even olive oil. I don’t buy a lot of rice and sugar. Once a month, if that, we get chicken or meat from my mother-in-law.”[308]

Bahia, 38, has four children and earns 6 JD per five-hour shift. She said she doesn’t receive any assistance, possibly because her husband is employed by the military, which disqualifies her from Takaful. “I used to be able to buy fruits, but now I can’t afford it anymore. We only eat three chickens per month.”

Tala, 35, just gave birth to her sixth child. She has been working since she was 16, she said, and earns 5 JD per four-hour shift. Her husband was also a day laborer but was injured on the job and can no longer work. “I have a one-and-a-half-month-old daughter…. My baby has been drinking chamomile tea for two days because I can’t afford milk,” Tala told Human Rights Watch. “I can no longer buy fruits, or any prepared food. I used to get chicken from [the charity] Tkiyet Um Ali but they cut me off to prioritize situations that are even worse than mine. There’s no work and there’s no help coming.”[309]

Under international human rights law, everyone has the right to access sufficient and adequate food needed to live a healthy and active life.[310] To protect this right, governments are obligated to enact policies and provide appropriate support to ensure that all people can afford safe and nutritious food at all times. 

Human Rights Watch was unable to find baseline data on food insecurity in Jordan prior to 2018, making it difficult to assess changes in food security over time, but a 2019 survey UNICEF conducted of Takaful applicants found that many struggle to afford nutritious food. According to the survey, one in three applicants have a borderline or low food consumption score, a standard measure that weighs both the frequency and nutritional value of food consumed in the previous seven days.[311] It also found applicants were more likely than the average Jordanian to rely on non-nutritional foods groups than on foods like vegetables, dairy products, and eggs, indicating that “on top of the 34 percent of families whose food consumption score is low or borderline, the average NAF Takaful applicant family’s diet is not balanced.” Even more worrisome, the report found inconsistency between survey responses and consumption patterns in one out of five respondents, with most consuming less than they reported. According to the report, “this could hint at the probability that at least 11 per cent of all families have a more severe food consumption score, or they are more vulnerable to food insecurity than reported.”

These stories illustrate the economic reality within which IMF-driven reforms occur. Rising prices exacerbated pre-existing economic hardship and the limited support from highly targeted social protection programs did not keep up. The expansion of emergency cash transfers under Takaful to respond to the pandemic appears to have broadened its reach to support many, although by no means all, people in need, but it was far from sufficient, and it has already ended even though the economic impacts of the pandemic are ongoing.

Electricity

Electricity is indispensable to the enjoyment of human rights, particularly the right to an adequate standard of living, and governments have an obligation to ensure that everyone in their territory has access to electricity that is sufficient, reliable, safe, clean, accessible, and affordable. Human Rights Watch has recognized that the right of everyone to an adequate standard of living includes the right of everyone to electricity.[312] The cost of electricity in Jordan is unusually high and many of the people interviewed in October 2021 described spending a significant share of their income on electricity. However, the rates they were paying were in fact cross-subsidized by a tariff imposed on some business sectors meant to decrease costs for households. Under the ongoing IMF program, Jordan reformed this cross-subsidized tariff structure and a new tariff structure went into effect in April 2022. The impact of the new tariff structure on households depends on usage amounts, with rates rising for some households and decreasing for others. The previous system, which had different pricing for seven consumption bands, was replaced with three bands under a new tariff structure for residential households:[313]

 

* a subsidy of 2-2.5 JD per month for usage below 600 kWh was announced alongside the tariff changes. See below for details.

** 1 fils is equivalent to 1/1000 dinar.

Jordanian authorities have presented the change as a reduction in tariffs and as only raising the rates on the wealthiest households, but the real picture is much more mixed.[314] Under the new system, households using less than 160 kWh per month will see their rates rise by 17 fils per kWh, while those using between 161 and 300 kWh per months will see their rates fall by 22 fils per kWh. According to the Energy Ministry, 58 percent of Jordanians use under 300 kWh/month, although it does not break it down further.[315] The same is true for the next band: 35 percent of households use between 300 and 600 kWh per month; those who use between 301 and 500 kWh will see their rates rise by 14 fils per kWh, while those who use between 501 and 600 will see their rates decrease by 14 fils. In the same vein, the 6 percent of households who use between 600 and 1000 kWh per month will see a moderate to steep rise in their tariffs, but households that use over 1,000 kWh per month—representing 1 percent of households and likely the wealthiest Jordanians—will see the steepest decline in their rates: a drop of 65 fils per kWh.

Studies show that higher electricity usage usually correlates with higher income, which is also reflected in how electricity usage rates are weighted in Takaful’s ranking system. However, this masks a much more complicated reality. Poorer households tend to be larger, rely on older, less efficient appliances, and be poorly insulated, making them more expensive to heat.[316] As Jordan’s National Social Protection Strategy notes, “whereas a third of the wealthiest decile are able to save on electricity bills thanks to owning solar water heaters, less than 4% of the poorest decile have such cost- saving assets.”[317]Nearly all of the low-income households Human Rights Watch interviewed in 2021 cited electricity bills indicating that they were using 500 to 600 kWh per month; well above the median usage of 300 kWh per month.

Based on the assumption that low-income households consume less electricity, government officials announced alongside the tariff changes a subsidy of 2.5 JD added to the bills of Jordanian households consuming less than 200 kWh per month and 2 JD added to households that use between 201 and 600 kWh.[318] Because the subsidy would make up for some or all of the increase, the government contends that 91 percent of Jordanian households would see no increase or a reduction.

However, it is unclear how long this subsidy will be in place. Also, the wording of the IMF loan agreement suggests that additional reforms may be anticipated.

In addition, most non-Jordanian citizens will not be eligible for the 2.5 and 2 JD subsidy, nor are they eligible for social protection to cushion any price increases. As the second IMF staff review of the loan program notes, they “instead have to rely on NGOs and international charities to provide assistance.” While UNHCR is funding a program that would provide six months of subsidy support to some Syrian refugees to help them cope with electricity price increases, it is far from meeting the need.[319] In 2018, before the economic ravages of the pandemic, more than 80 percent of Syrian refugees lived below the Jordanian poverty line, and funding for organizations providing assistance to refugees has declined dramatically in recent years.[320] A survey conducted by UNHCR in 2020 found that refugees in Jordan spend about 30 percent of their income on water, electricity, and gas bills, and another one conducted in mid-2021 found that 68 percent of Syrian refugees in Jordan had seen their incomes decrease since the beginning of the pandemic.[321]

The IMF acknowledged that tariff reform would need to be offset by increased investment in social protection, but promoted a “well-targeted” program:

Measures under the plan—including elimination of large cross subsidies and implementation of the new tariff-adjustment mechanism—should be implemented as swiftly as possible, and complemented by a well-targeted social protection scheme to safeguard low-income and vulnerable households.[322]

This is in line with a 2017 study commissioned by UNICEF that projected the impact of subsidy reform in Jordan and found “that the reform alone [that is, without cash transfers] would further exacerbate poverty through increased consumer prices.”[323]

These reforms come at a time when a large number of Jordanians say they have experienced an increase in their electricity bills even though the tariff has remained unchanged.[324] The reason for the increase is unclear, although some government officials have acknowledged the complaints.[325] Everyone with whom Human Rights Watch spoke described skyrocketing electricity bills in recent years, with nearly everyone struggling to pay each month even when using the bare minimum number of appliances. Aiman Albatayneh, an energy engineer and assistant professor at the German Jordanian University who recently published a study on Jordan’s electricity sector, told Human Rights Watch he believes the increase is due to climate change contributing to more hot and cold extremes, leading people to purchase or rely more on air conditioning and heating.[326]

The cost of electricity consumed an enormous part of people’s income and they said they were often simply unable to pay their bills. A 40 JD bill is roughly 15 percent of a minimum monthly wage, and day laborers earn far less than that.

“I used to have an air conditioner, freezer, grill and the bill was 6 JD,” Ibrahim told Human Rights Watch. “Now it’s around 35 JD…The air conditioner is broken and the whole house has LED lights, a fridge, and a water cooler.” He said that in around 2019, the electricity company cut off his electricity for six months because he couldn’t pay his bills. As noted, the only assistance he has received was under the temporary Covid-19 expansion of Takaful.[327]

Khaled, a driver, who is the only person in his family who is working after his father passed away, said he saw his electricity bills double from 20 JD to 40 JD in recent years. He said he didn’t apply for assistance because he never received a message about it and didn’t know how to apply.[328] Bassam has seven kids and his wife doesn’t work, but he was denied support because he has a restaurant registered in his name although it has since closed down. “The price of electricity increased three times over two years. I used to pay 15 JD, now I pay 50. No air conditioner, just basic stuff. The water heater costs the most.” He can’t pay his bills so he plays cat and mouse with the electric company. “We’re used to it. They cut off electricity, they turn it off, we go beg for money, and get it turned back on. After three months of not paying, it’s cut off. We sit in candlelight and don’t use food that needs refrigeration.”[329]

Human Rights Watch visited the office of an electrical company in Marka that manages shut-offs and a representative explained that “the rule is that after three months if an unpaid bill is more than 100 JD, the administration asks us to cut off power.” He expressed sympathy with such customers, noting that he is “one of them” but said after a while he is left with no choice. “We cut off around 200 families per week. Usually they come pay and we turn back on. Sometimes it’s longer. Sometimes it’s because they’ve left the house. But sometimes it’s because they don’t have money and they sit in the dark.”

A vegetable seller in Jabal al-Hadid said he is paid around 150 JD per month, although his salary is supposed to be higher. His family receives around 125 JD through Takaful every three months, although it still isn’t enough to manage rising electricity bills, along with their other expenses.

“We have a hard time paying bills. Electricity is now 40-45 JD, sometimes for two to three months we are unable to pay. They cut it and then we pay half and they turn it back on.”

People are turning to private charities to help them pay their electricity bills, a representative from the charity in al-Quaissmeh told Human Rights Watch. There is “a significant increase in the number of people who are bringing them their bills so their electricity doesn’t get cut off. But they don’t fit our strict criteria so we can’t help them. The situation is very difficult.”

While it is unclear what is driving these increases, it highlights the importance of adequate social protection systems to address energy poverty, particularly ahead of reforms that can affect affordability.

In Jordan, the Ministry of Social Development and the National Aid Fund developed a scheme to install solar panels on poor households that are registered with the NAF, and to whom specific conditions apply, to potentially offset price increases from the removal of cross-subsidies.[330] The program was beset with challenges, given the difficulties with registration and application of conditionalities. According to media reports, over 2,000 households benefited from the program as of September 2020.[331] In 2021, the program was set to benefit an additional 7,000 families.

The Adequacy of Takaful

In some ways, Takaful marks an important advancement in Jordan’s social protection system. Nearly everyone with whom Human Rights Watch spoke had heard of the program and most had applied, although many encountered difficulties doing so. Many had benefited from it or the National Aid Fund’s cash transfer program. An early report conducted in July 2020 by UNICEF found that the program has reduced poverty rates among beneficiary families by as much as 7 percent.[332] Yet the interviews also surfaced the limitations of a highly targeted social protection program, particularly as a means of offsetting price increases that affect an entire population.

This report focuses primarily on limitations that arise from the scope of the program, investigating its adequacy even if it works as intended. However, in reality targeting rarely works as intended and is often beset by large inclusion and exclusion errors, administrative hurdles, and other issues. The World Bank and others have promoted the data-intensive algorithmic systems to select beneficiaries as a way of addressing these problems, but separate Human Rights Watch research investigating the design and accessibility of Takaful has found that many of these problems persist and some new ones are created.[333] Many people in Jordan do not qualify for cash transfers because their hardships do not fit the targeting algorithm’s faulty model of poverty. But even if these problems were to be resolved, the narrow scope of the program is inadequate to protect rights.

Categorical Exclusions

Only Jordanian citizens are eligible for Takaful, yet the country’s large refugee population is also impacted by the price increases and is on average poorer than Jordanians, making it harder for them to absorb the higher costs. As noted, planned electricity reforms in April 2022 would further drive up costs of electricity for non-citizens by removing all subsidies for them. Under international human rights law, governments may not discriminate against non-citizens in the distribution of benefits. Over 80 percent of Syrian households in Jordan were either food insecure or vulnerable to food insecurity in 2018, a period before both significant price increases and the Covid-19 pandemic.[334] Among non-Syrian refugees, 66 percent were food insecure or vulnerable to food insecurity.[335]

The UNHCR and UNRWA, an agency that supports Palestinian refugees, cooperate with the government to develop specific programs that provide assistance to Syrian and Palestinian refugees to offset the cost of these price increases, but in reality, such programs are capped in number and funding challenges have meant that the programs are often shut down prematurely. A Syrian refugee who works in a supermarket in Marka told Human Rights Watch that he doesn’t even get the money to which he is entitled. “I am entitled to 20 JD per month, but I don’t get it. They came for a visit and said in my situation I should, but I still don’t. And they didn’t raise the amount with prices.”

A recent IMF review of the program appears to acknowledge these limitations, noting the government’s decision to “fix bread prices until end-2023, given their importance to mitigating the impact of high global food prices on the population and the inability of the current social safety schemes to reach all vulnerable groups, including refugees.”[336]

Moreover, unlike Jordanian men, Jordanian women are unable to pass along their citizenship to their spouse or children.[337] As a result, if a Jordanian woman who is married to a foreigner applies to Takaful, if eligible, she will receive benefits as though she were a one-person household, which is 40 JD per month.[338]

Scope of Coverage

Takaful was intended to provide cash assistance to 85,000 households; it has since been expanded to provide 120,000 households with up to 136 JD per month. While an improvement in Jordan’s safety net, it leaves the vast majority of families in poverty with no support as they absorb the price increases associated with IMF-driven reforms. As noted, some 2.7 million Jordanians—approximately 557,500 households, based on an average household size of 4.8 people—currently live below the official poverty line, but even this likely does not come close to capturing the true number of people who struggle to realize their rights, particularly in the context of rising prices.

Practically speaking, this means that the targeting creates situations where people with even trivial assets may not qualify for support. Many people complained that they believed a car registered in their name pushed their ranking too high to receive benefit from the program. NAF clarified that families with cars that are less than five years old are excluded from the program, while the algorithm weighs cars older than five years old based on factors such as the car’s value in determining the family’s poverty ranking.[339] In other words, families with older cars could be less likely to qualify for Takaful compared to families without, all else being equal. But, as one person described his situation, owning a car is not a reliable litmus test for need. He said: “Okay I have a car but I can’t always fill it with fuel. There isn’t always electricity or water so sometimes I have to go work without washing up.”[340] A woman with two children whose husband is an agricultural worker described how she was unable to send her son to school because she can’t afford the transportation or cost of books. She cited the old car they sometimes use to transport water and firewood as a possible factor in their exclusion from Takaful:

             

The car destroyed us. Look at it, it’s parked outside.… Sometimes we don’t have the money to fill it up with diesel, so we walk to the street and wait for someone to pass by and agree to pick us up.[341] 

Other interviewees noted other assets they suspected led them to be rejected from the program. A vegetable seller said he believed he was rejected because of a piece of land he had inherited in the countryside. “We have to live in a tent for them to accept us,” he complained.[342]

At the same time, because the program is intended for the working poor, people who do not have any income are not eligible. There are some NAF programs that may provide them with assistance, for example for older persons and orphans, but these cover only a small number of households and benefit amounts are typically very small.[343] One person admitted to Human Rights Watch that he fabricated a salary out of desperation, but this is a decision fraught with risk. At the same time, several said they needed to artificially inflate their salaries or reduce their expenses because the application does not accept expense figures that generally exceed twenty percent of their income.[344] But this does not reflect the reality of their spending, people said, where they rely on a combination of income, one-off gifts from family, and other ad hoc sources to make up for significant shortfalls.

Even if the program were expanded to cover everyone living under the poverty line it would miss many families who are struggling since the poverty line is set so low. And research has found that targeted programs that only cover those in poverty create a “missing middle” effect, where the people near poverty are left worse off than those in poverty who are supported by the program.[345] In addition to leaving many people in need unsupported, this fuels social tensions around the programs and erodes public support for them.

A UNICEF survey highlights the thin line that separates families who receive support from those who don’t. While it found that Takaful beneficiaries relied most on negative coping strategies to afford food, the difference between them and eligible non-beneficiaries (that is, households under the poverty line but less poor than beneficiaries) and ineligible applicant households (that is, applicant households above the poverty line) was marginal. For example, 76 percent of beneficiary families said they reduced their numbers of meals per day, while 71 percent of both eligible non-beneficiary families and ineligible families said they did the same.[346] Similarly, 79 percent of beneficiary families said they reduced portion size, and the same was true for 76 percent of eligible non-beneficiary families and 70 percent of ineligible families.[347] In other words, seven out of ten families who applied to Takaful but are not eligible because they earn more than the poverty line said they reduce the number of meals they eat and their portion sizes in order to afford food, but even if Takaful were expanded to include all households below the poverty line they would still not qualify.

Moreover, these numbers don’t capture the impact on people who did not apply because they did not believe they would qualify, yet still need to make painful cuts to their standard of living due to rising prices. Samer, who owns a water shop that has been struggling due to rising cost of living but still earns significantly more than the poverty line, said he transferred his three kids to a cheaper school to save on tuition costs. “We stopped going places because of the cost of fuel. We used to go out, eat out. Now everything is limited,” he said. People like Samer who are impacted not only because the price of goods is higher for them but because people have less money to spend at their shops exemplify the “fiscal multiplier effect” that the IMF blamed for why its optimistic predictions turned out to be disastrously wrong following the Financial Recession. Notably, the IMF’s guidance note on how to operationalize engagement on social spending in the wake of the pandemic cautions staff that when making projections to assess the potential impact of it reforms, they “should also be mindful of the positive impact of social spending on growth, via potentially high fiscal multipliers.”[348]

At the same time, because Takaful provides a maximum disbursement of 136 JD per month, it still leaves many struggling to meet their basic needs. Rabia, an agricultural worker with four children, said she receives 160 JD every three months.[349] “It’s not enough at all. We’re trying to work very hard, but there are gaps,” she says. “We don’t buy anything, not even olive oil. I don’t buy a lot of rice and sugar. Once a month, if that, we get chicken or meat from my mother-in-law.”[350]

Funding

IMF reforms are intended to be permanent but social protection programs need funding to remain operable. There is always a risk of funding cuts, but research indicates that targeted programs are especially vulnerable since, by definition, they benefit a constituency that is often the most politically disenfranchised.[351]

In Jordan, the risks of funding cuts may be especially high since funding for both the cash transfers and operation of Takaful has largely depended on time-limited external funds. A World Bank project provided funding for Takaful’s original $200 million three-year budget, and UNICEF’s design and implementation of the program has been funded by various donors. The expansion of the program during Covid-19 did rely on a combination of government and donor funds, including from the World Bank, but this was a temporary allocation.

In this context, it is especially worrisome that in 2020, prior to the pandemic and the infusion of new emergency resources into Takaful, Jordan was on track to underfund the program by at least a third. In 2020, the budget called for 65 million JD to support the program, but, according to World Bank documents, “due to fiscal pressures, the [Government of Jordan] was only able to allocate JOD 45 million to Takaful in 2020, leaving a funding gap of JOD 20 million.” If authorities chose to cut the budget even when it was funded by a World Bank project, it raises serious concerns about its commitment to funding the program from the general budget, particularly given the fiscal pressures generated by the pandemic. A former senior government official told Human Rights Watch that “unless there is a shift in public spending, I don’t see [Takaful] being sustainable – meanwhile populations are growing and unemployment is increasing.”

Social Protection is Not a Panacea

This report focuses on the limitations of excessively low social protection floors, which are tied to highly targeted cash transfer programs, to offset the impacts of IMF-driven adjustments for people on low incomes because that is the primary tool the IMF emphasizes. However, social protection alone is not sufficient to make economies that fulfill everyone’s economic and social rights.

Jordan has recognized some economic and social rights, such as the rights to education and to work in its constitution.[352] Its Comprehensive National Plan for Human Rights for 2016-25 includes a commitment to improving the rights to work, education and health but very little has been implemented so far. Moreover, Jordanian laws do not specifically recognize the right to social security or the right to electricity. Ahmad Awad, the head of the Phenix Center, an independent organization promoting sustainable development, emphasized the importance of labor protections, including a living minimum wage and freedom of association, as critical for reducing poverty and inequality. As already noted, Jordan’s minimum wage is far below a living wage and hourly wage-earners are excluded. The law also imposes several restrictions on freedom of association.[353] Yet few IMF programs include reforms that strengthen labor protection.

The absence of any such reforms in IMF-driven programs is particularly stark in the case of Jordan, where labor rights are sorely lacking. For instance, in 2022, the Global Sustainable Development Report issued by the UN scored Jordan as 0.5 of 1 in “fundamental labor rights,” a decrease from 0.57 in 2014.[354] More than 40 percent of workers are in the informal economy, according to the Jordan Strategy Forum, making them particularly vulnerable to abuse. Moreover, attacks on independent labor unions have only grown throughout the last decade. For example, in July 2020, authorities outlawed and dissolved the Jordan Teachers’ Syndicate, a union that represents 140,000 teachers, after it campaigned to secure wage increases to which the government had committed in 2019 following a nationwide teachers’ strike.[355] Yet none of the IMF programs make reference to freedom of association.

According to Ahmad Awad, this escalation in attacks is a culmination of four decades of eroding labor rights in Jordan. “During the last 40 years, we have become more and more unbalanced – the pressure on wages and freedom of association ha[s] increased,” he told Human Rights Watch. There are more than 120 employer associations, but only 17 recognized workers’ unions, which are unified by law into a government federation that is not governed by transparent or democratic principles. The Minister of Labor holds the power to dissolve any union, which is especially problematic, Awad said, because labor ministers are generally drawn from the business elite. “There is a revolving door between the private sector and ministries, so no real division between the two.”

Mithkal Zinati, the head of a Farmer’s union with 850 members, said the government rejected his union’s bid for recognition, and he is now appealing the decision. He organizes farmworkers who are especially vulnerable to exploitation, in part because they are not covered by Jordan’s minimum wage since it sets a monthly minimum which effectively excludes hourly workers. In practice, farmworkers are paid 1.25 JD per hour, he said, which itself is the result of workers organizing to increase the hourly wage from 1 JD. At 1.25 per hour, even if workers worked 40 hours per week, their monthly income would be 200 JD, which is significantly less than the 260 JD minimum wage. But in reality, many often struggle to find daily work. “The economic problems can’t be solved until there is the ability to organize, which is under attack in this country,” Zinati said.[356]

Rather than address these fundamental labor issues, the IMF’s program has focused on increasing employment by reducing costs for businesses, including at the expense of workers. The 2016 program included a structural benchmark that allowed certain employers to reduce their social security contributions for “youth.” The 2020 program doubled down on this approach and expanded the reform to include all start-up companies.

Labor protections are themselves but one aspect of a broader framework of policies necessary for human rights-aligned economies, along with quality public services, universal social protection, inclusive development policies, environmental protections, and adequate regulation of the private sector, among others. If the IMF is serious about building resilient, sustainable, and inclusive economies, it should begin with these principles as it shapes its programs.

 

IV. International Human Rights Standards

The International Covenant of Economic, Social and Cultural Rights (ICESCR) enshrines all people’s rights to, among other rights, the highest attainable standard of health, education, social protection, just and favorable conditions of work including a living wage, and an adequate standard of living, which includes the rights to food, housing, and water. Human Rights Watch has found that this right also includes a right to electricity. While the full realization of these rights is expected to be realized progressively, the covenant requires governments to immediately:

Take steps, individually and through international assistance and cooperation, especially economic and technical, to the maximum of its available resources, with a view to achieving progressively the full realization of the rights recognized in the present Covenant by all appropriate means, including particularly the adoption of legislative measures.[357]

By setting limits on government spending and imposing economic and fiscal reforms, the IMF plays an important role in shaping governments’ ability to fulfill their human rights obligations, including with regard to the rights to social security and to an adequate standard of living. As such, it has an obligation enable, and not to impede, governments from being able to fulfill these obligations.[358]

The UN Committee on Economic, Social and Cultural Rights (CESCR), charged with interpreting the ICESCR, notes that international financial institutions (IFIs) “are obligated to comply with human rights” and explicitly disagrees with the IMF’s position that its Articles of Agreement preclude it from doing so.[359] It explicitly rejects the IMF’s argument that applying international human rights law would be inconsistent with its Articles of Agreement:

In discharging their duty to comply with human rights under international law, international institutions are not exercising powers that they do not have, nor are they taking into account considerations they would be obliged to ignore based on their statutes; rather, it is in the exercise of the powers that have been delegated to them by their member States that they should refrain from adopting measures that would result in human rights violations.[360]

The covenant also obligates ratifying governments, where needed, to facilitate “international assistance and coordination” to protect economic, social and cultural rights. Particularly in situations of economic crisis, humanitarian aid and debt relief, technical cooperation, and international concerted action are often critical to enabling governments to adequately respond to protect rights.[361]

Austerity and Human Rights

How lenders and governments respond to economic crises has enormous impacts on rights in both the short and long term. As the CESCR notes in General Comment 4:

[T]he obligations under the Covenant continue to apply and are perhaps even more pertinent during times of economic contraction. It would thus appear to the Committee that a general decline in living and housing conditions, directly attributable to policy and legislative decisions by States parties, and in the absence of accompanying compensatory measures, would be inconsistent with the obligations under the Covenant.

Numerous human rights bodies and experts have especially emphasized the risks that austerity poses to the enjoyment of human rights. In 2013, the UN Office of the High Commissioner for Human Rights (OHCHR) issued a report pursuant to a UN General Assembly resolution that found that the implementation of austerity following the 2008 Financial Recession, often as the result of IMF conditionalities, delayed recovery in many countries and that “the ability of individuals to exercise their human rights, and that of States to fulfil their obligations to protect those rights, has been diminished.”[362] It noted both the direct impact of spending cuts on people who rely on publicly supported goods and services, such as health, education, and social protection, as well as the indirect impacts of austerity, such as on employment.[363]

These impacts fall disproportionately on women, migrants, and older persons, the report found, because they are often in more precarious work and are especially vulnerable to decreases in wages.[364] Human rights organizations, including Human Rights Watch, Amnesty International, and the Center for Economic and Social Rights, have documented the human rights impacts of austerity, as have many other research groups, such as Oxfam, academics, and other research institutions.[365]

While international human rights law recognizes that public spending cuts may in certain situations be necessary, there are strict rules for when and how these may be pursued to ensure that they avoid harming human rights, including that they “demonstrate that all other alternatives have been exhausted, and provide safeguards to protect human rights, particularly the rights of the most vulnerable.”[366] Austerity measures must also be temporary, necessary, proportionate, respectful of minimum core human rights obligations, and non-discriminatory. To ensure these conditions are met, states are expected to submit prospective polices to the public for broad scrutiny and inclusive consultation, particularly by those most likely to be affected.

The report’s guidance became the basis for a CESCR statement, published in July 2016, that reiterates the strict conditions set out in the OHCHR report, as well as financial lenders’ international human rights obligation to ensure that conditionalities do not cause governments to violate their respective rights obligations. [367] To that end, it obligates both lenders and governments to conduct human rights impact assessments “prior to the provision of the loan, in order to ensure that the conditionalities do not disproportionately affect economic, social and cultural rights nor lead to discrimination.”[368] The statement builds on two sets of guiding principles related to human rights and economic policy considerations, both adopted by the UN Human Rights Council in 2011.[369] The first, on foreign debt, sets out a process for creditors, lenders, and states to ensure that debt servicing does not come at the expense of human rights. Recognizing the burden debt places on governments’ ability to invest in rights, including due to austerity measures, it calls for debt restructuring and relief where appropriate to alleviate the burden, and on financial institutions not to impose conditions that negatively impact human rights. The second, on extreme poverty, enumerates how states and financial institutions must pursue economic policies that reduce poverty and progressively realize economic and social rights. For example, “Fiscal policies, including in relation to revenue collection, budget allocations and expenditure, must comply with human rights standards and principles, in particular equality and non-discrimination.” Both sets of guiding principles require states and financial institutions to undertake human rights impact assessments and engage in transparent and credible public dialogue to ensure policy choices are in line with human rights obligations.

In March 2019, the Human Rights Council adopted a third set of principles to guide states and financial institutions in their response to economic crises, developed by Juan Pablo Bohoslavsky, the UN Independent Expert on the effects of foreign debt and other related international financial obligations of States on the full enjoyment of human rights, particularly economic, social and cultural rights. These principles are intended to provide lenders and governments with a framework for responding to economic crises in ways that protect and advance rights in both the short and long term: “Above all, Governments must ensure that measures put in place serve to pursue economic recovery for the benefit of the whole population, instead of only a few.”[370]

To achieve this, governments and lenders are expected to conduct human rights impact assessments “that demonstrate that their proposed economic reform measures will realize, not undermine” human rights.[371] The assessments, which should be conducted in a transparent, participatory process and then published, should identify and address any human rights harms stemming from their policy choices.

In assessing possible policy choices, states must consider several things. First, economic reforms must not be discriminatory. To prevent discriminatory impacts, the assessments “should seek to identify and address the potential and cumulative impacts of measures on specific individuals and groups and protect them from such impacts.”[372] Discriminatory impacts on women are of particular concern, and reforms should not only prevent such discrimination but affirmatively “promote substantive and transformative gender equality.”[373] The principles caution that:

Economic reforms which encourage, among other things, labour market flexibilization, reductions in the coverage of social protection benefits and services, cuts to public sector jobs and the privatization of services tend to have a negative impact on women's enjoyment of human rights.[374]

They also note that disenfranchised and marginalized groups are often disproportionately impacted by certain economic reforms and “[d]irect, indirect, multiple and intersectional discrimination… needs to be carefully assessed and prevented.”[375]

Second, how states generate and allocate resources is key to effectively responding to an economic crisis in ways that protect, respect, and fulfill rights. Under the guiding principles, states must not only use existing resources, but also “generate potential resources in a sustainable way,” including through seeking international aid and assistance, tackling tax evasion, and raising taxes in a progressive manner. Other human rights bodies have also found fiscal policy, in particular raising taxes progressively, to be essential to states’ ability to fulfill their human rights obligations. For example, the CESCR notes in General Comment 24 that:

The obligation to fulfil requires States parties to take necessary steps, to the maximum of their available resources, to facilitate and promote the enjoyment of Covenant rights, and, in certain cases, to directly provide goods and services essential to such enjoyment. Discharging such duties may require the mobilization of resources by the State, including by enforcing progressive taxation schemes.[376]

Third, any measures that would result in retrogression—such as an increase in prices or decrease in support or wages that hurts the ability of people on low incomes to enjoy their rights—is “considered a prima facie violation” of rights, and is only permissible if there are no other policy alternatives that would be less harmful to rights.[377] In considering options, the guidelines note that states should utilize economic, fiscal, monetary, and social policies to avoid austerity. In particular, they note the important role of counter-cyclical spending: “Fiscal policy also has a clear and well-documented effect on economic growth: while pro-cyclical contractionary fiscal policy reduces economic growth, expansionary fiscal policy during downturns increases growth.”

Where governments are able to demonstrate they are unable to avoid measures that lead to retrogression, such measures must be temporary, proportionate, non-discriminatory, and protect the minimum core obligations of economic and social rights. They also must be based on transparency and the genuine participation of affected groups and subject to meaningful review and accountability procedures, including human rights impact assessments. This is consistent with the position of other human rights bodies and experts, including the CESCR’s July 2016 statement on austerity and human rights.

Bohoslavsky published an article in Columbia Law School’s Human Rights Law Review in 2021 making the case that there is “a solid legal basis to make the case for a prima facie inconsistency between the imposition of austerity policies in times of recession and the enjoyment of human rights,” and that “international financial institutions may be held responsible for complicity in the imposition of economic reforms that violate human rights.”[378] In October 2020, Human Rights Watch joined a call by 504 civil society organizations that called on the fund to “immediately stop promoting austerity around the world, and instead advocate policies that advance gender justice, reduce inequality, and decisively put people and planet first.”[379]

This report finds the IMF is not adequately exploring alternative options to austerity measures that harm rights. Moreover, it does not adequately justify the reliance on these measures or assess their rights impact on people on low incomes, including women and marginalized groups. While the lack of information and other factors prevents a proper analysis of whether social spending floors adequately mitigate such impacts in all cases, IMF guidance and practices, particularly with regard to targeting, as well as evidence from Jordan, raise serious concerns that they do not.

Right to Social Security

Both “social protection” and “social security” describe a range of policies and programs premised on the principle that everyone should enjoy all their economic, social and cultural rights at all stages of their lives, no matter the circumstances into which they are born or the crises or challenges they may face.

While the term “social protection” has become popular within parts of the United Nations and some international development organizations, there is no uniformly accepted definition for what it entails, whereas the term “social security” is clearly described in international human rights law as a set of individual entitlements that protect against income insecurity throughout people’s lives, including during common life events, such as older age, unemployment, sickness, or birthing and caring for dependents.

Article 22 of the Universal Declaration of Human Rights of 1948, for example, spells out the essential elements of the right:

Everyone, as a member of society, has the right to social security and is entitled to realization, through national effort and international co-operation and in accordance with the organization and resources of each State, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality.[380]

Since then, the right to social security has been widely enshrined in countries’ national constitutions and reinforced through a range of other international conventions and frameworks, including the ICESCR.[381] The CESCR for example, defines this right to encompass at least nine areas of support: health care; sickness; older age; unemployment; employment injury; family and child support; maternity; disability; and survivors and orphans following the death of a breadwinner.[382]

States that are parties to the ICESCR assume the obligations to respect, protect, and fulfill the right to social security along each of these areas of support, including by making these programs available, accessible, acceptable, and adaptable.[383] This also requires providing benefits, whether in cash or in kind, that are adequate in both amount and duration.

As with all other human rights, governments need also to realize the right to social security without discrimination on the grounds of gender, age, disability, race, nationality or immigration, or other status. This means that countries should be careful to ensure that the design and operation of social security systems do not directly or indirectly discriminate against anyone, such as through language or technology barriers that can cause de facto exclusion or negative treatment. As with other human rights, the right to social security should be enshrined in domestic law, and give victims of violations an effective remedy.

More recent international human rights law instruments recognize a right to social protection, in addition to and as distinct from, the right to social security. For example, a recent protocol to the African Charter on Human and Peoples’ Rights enshrines both rights separately.[384] According to the Protocol, social security protects against income insecurity caused by events such as unemployment, sickness, or maternity. And social protection encompasses all forms of social security while also including strategies and programs that help ensure a minimum standard of livelihood, health services, and care.

In this sense, social protection corresponds to a set of policies and programs that governments need to put in place to fulfill their obligations to realize a range of economic, social and cultural rights under all circumstances, such as the rights to education, health, and an adequate standard of living, which includes the rights to food, housing, water, and sanitation, among others.

Similarly, the concept of the “social protection floor,” developed by the CESCR and the ILO, draws from the human right to social security, comprising at least four social security guarantees:[385]

1. Access to a nationally defined set of goods and services, constituting essential health care, including maternity care, that meets the criteria of availability, accessibility, acceptability, and quality;
2. Basic income security for children, at least at a nationally defined minimum level, providing access to nutrition, education, care, and any other necessary goods and services;
3. Basic income security during people’s working life, at least at a nationally defined minimum level, for people who, including due to social and attitudinal barriers, are unable to earn sufficient income, or have extra living costs in particular in cases of sickness, unemployment, maternity, and disability; and,
4. Basic income security, at least at a nationally defined minimum level, for older people.

Right to an Adequate Standard of Living

Food

The Universal Declaration of Human Rights (UDHR), places the right to food within the broader concept of a “right to an adequate standard of living”:

Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.[386]

The International Covenant on Economic and Social Rights is the fundamental international treaty governing socioeconomic rights. Article 11 of the ICESCR sets out the concept of an “adequate standard of living”:

The States Parties to the present Covenant recognize the right of everyone to an adequate standard of living for himself and his family, including adequate food, clothing and housing, and to the continuous improvement of living conditions. The States Parties will take appropriate steps to ensure the realization of this right, recognizing to this effect the essential importance of international co-operation based on free consent.[387]

The ICESCR specifically obliges states to ensure “the fundamental right of everyone to be free from hunger”, including through measures such as specific programs on food “production, conservation and distribution”.[388]

The UN Committee on Economic, Social and Cultural Rights (CESCR)—the body of independent experts established to monitor the implementation by states of the ICESCR and to provide authoritative interpretation of the specific rights in the covenant—has also offered its guidance setting out what the “right to food” within the ICESCR means and clarifying what duties states have. According to the CESCR:

The right to adequate food is realized when every man, woman and child, alone or in community with others, have physical and economic access at all times to adequate food or means for its procurement. The right to adequate food shall therefore not be interpreted in a narrow or restrictive sense which equates it with a minimum package of calories, proteins and other specific nutrients.[389]

The core content of the right to adequate food implies:

The availability of food in a quantity and quality sufficient to satisfy the dietary needs of individuals, free from adverse substances, and acceptable within a given culture; The accessibility of such food in ways that are sustainable and that do not interfere with the enjoyment of other human rights.[390]

The CESCR also sets out clearly how the right to food is inextricably linked to the broader requirement of states to eradicate poverty in order to ensure everyone can enjoy all their human rights:

… the right to adequate food is indivisibly linked to the inherent dignity of the human person and is indispensable for the fulfilment of other human rights enshrined in the International Bill of Human Rights. It is also inseparable from social justice, requiring the adoption of appropriate economic, environmental and social policies, at both the national and international levels, oriented to the eradication of poverty and the fulfilment of all human rights for all.[391]

The current United Nations Special Rapporteur on the right to food defines the right as “the right to have regular, permanent and unrestricted access, either directly or by means of financial purchases, to quantitatively and qualitatively adequate and sufficient food corresponding to the cultural traditions of the people to which the consumer belongs, and which ensure a physical and mental, individual and collective, fulfilling and dignified life free of fear.”[392]

The Office of the High Commissioner of Human Rights has further emphasized that:

“The right to food is not a right to be fed, but primarily the right to feed oneself in dignity.... The right to food requires States to provide an enabling environment in which people can use their full potential to produce or procure adequate food for themselves and their families.”[393]

The Convention on the Rights of Persons with Disabilities (CRPD) establishes the right to an adequate standard of living in similar terms as the treaties discussed above, however it adds new content and a specific disability focus to this right. CRPD article 28.1 underscores that people with disabilities have a right to an adequate standard of living for themselves and their families, “including food, clothing and housing, and to the continuous improvement of living conditions.”[394] The CRPD obliges states to take steps to safeguard and promote the realization of equal access to water services and to appropriate and affordable services, devices, and other forms of assistance that are needed because of a disability as well as access to social protection and poverty reduction programs. These programs are especially important for women and girls.[395]

Electricity

Human Rights Watch recognizes that the internationally protected right to an adequate standard of living includes the right of everyone, without discrimination, to sufficient, reliable, safe, clean, accessible, and affordable electricity. Access to electricity is critical to ensuring other basic rights, including but not limited to the rights to health, housing, water, sanitation, and education.

States, as well as de facto authorities and occupying powers, have a duty to ensure that everyone in their territory or jurisdiction has access to electricity. This means ensuring adequate and sustainable electricity generation, supply networks, and international cooperation to ensure reliable electricity access for everyone.

States should take all necessary action to ensure everyone’s access to energy is met through renewable supplies of electricity, and pursue through laws, policies and other measures, practices that minimize the impact on the environment and human rights of electricity production, supply, and use.

 

Acknowledgments

This report was written by Sarah Saadoun, senior researcher and advocate on poverty and inequality in Human Rights Watch’s Economic Justice and Rights (EJR) division. Sara Kayyali, former senior researcher in the Middle East and North Africa (MENA) division, provided extensive research in Jordan.

The report was edited by Arvind Ganesan and Sylvain Aubry, director and deputy director of the EJR division, respectively; Lena Simet, senior researcher and advocate on poverty and inequality in the EJR division; Adam Coogle, deputy director in the MENA division; Maria McFarland Sánchez-Moreno, acting deputy program director; and Clive Baldwin, senior legal advisor. Numerous specialist reviewers provided invaluable contributions, including Katharina Rall, senior researcher in the Environment and Human Rights Division; Juan Pappier and Martina Rapido Ragozzino, acting deputy director and senior research assistant in the Americas division, respectively; Amos Toh, senior researcher in the Technology and Human Rights Division; and Carlos Rios-Espinosa, associate director of the Disability Rights Division. Additional editorial assistance was provided by Jack Spehn, senior associate in the EJR division, who also prepared this report for publication.

Human Rights Watch is deeply grateful to the people who shared their experiences coping with rising living costs in Jordan. We also appreciate the many people who shared their time and expertise to illuminate and refine our understanding of the issues included in this report. In particular, we would like to thank Isabel Ortiz, director of the Global Social Justice Program at the Initiative for Policy Dialogue, based at Columbia University in New York, for reviewing the report, and Allison Corkery, director of Rights Claiming and Accountability at the Center for Economic and Social Rights, for reviewing portions of it.