The IMF on Protecting the Poor during Fiscal “Consolidations”: Better Late Than Never

October 23, 2017

Inequality and inclusive growth were high on the agenda of the Annual Meetings of the International Monetary Fund and World Bank earlier this month. We are glad about that, but the under-reported story here is that this prominence marks a dramatic shift in the IMF over the last two decades in the IMF’s approach to the relevant challenges for the poorest countries, including on the issue of social safety nets and social expenditures.

The IMF is waking up on how to help countries manage fiscal reforms AND protect the poor

In the 1990s and early 2000s, the IMF was still largely living up to its nickname “It’s Mostly Fiscal.” Distributional matters were considered a matter of politics and under the purview of national governments—and for analytic and policy work a matter for the World Bank. (In 2002 a report of the IMF’s Independent Evaluation Office recommended that during Article IV consultations, staff ask whether authorities had identified social programs they would like to protect in the event of a crisis. Several Board members resisted that recommendations, viewing it as “impractical”—and infringing on World Bank responsibilities.) And as recently as 2007, this CGD report (led by the former head of the IMF’s IEO) found that IMF programs in African countries, in part because they systematically understated expected future donor support for health programs, made recommendations for fiscal consolidation that were not necessary, and for reductions in civil service numbers that ended up with governments, faced with conditionality on civil service spending cuts, hitting the health sector (and possibly the education sectors) hard.

But with the global financial crisis, the failure of harsh bouts of fiscal austerity in Greece to restore growth, and the growing body of IMF research income distribution and its causes and consequences, the IMF has changed. In its official statements, the institution now recognizes explicitly the potential trade-off between fiscal “consolidation,” for all its merits in many circumstances, and the effects on the poor of resulting cuts in social spending. Those cuts not only reduce spending on basic schooling and health care (the obvious concern of the IMF and fiscal authorities), but if they lead to pension reductions and civil service job eliminations, can move back into poverty incipient middle class “strugglers,” as well. 

In spring of this year, for example, the Fund released this Staff Paper, reporting on research indicating that in the poorest countries eligible for grants and low-cost Fund loans, IMF programs since 2010 have included “floors” or “indicative targets” on education and health spending and on whatever social safety net programs borrowing countries had (see Broome for the provenance of this new “floor” approach), and that in two thirds of the 68 loan programs examined, the resulting “indicative conditions” were met.

As Jo-Marie Griesgraber of New Rules for Global Finance notes, the paper on which the Staff Report is based is at least transparent (e.g. on the failed “floor” in one-third of borrowing countries; “in every classroom in the world, 66% is still failing”), and we would add that the paper, now in the form of a Staff Report is on the record, clarifying that IMF country operations staff have a responsibility to take note.

The limits of the IMF’s indicative targets on social protection

At the same time, there is a long way to go. The underlying research in the recent report was limited in its aims, definition, and scope. The main conclusion is that IMF programs in poor countries have leaned, since 2010, in the right direction—by minimizing cuts in those programs most likely to reach the poor. But that conclusion relies heavily on the expenditure side, e.g., on countries’ maintaining floors during IMF program periods on total expenditures on health and education and social transfers as a share of the countries’ GDP.

But indicative targets on spending on cash transfers, education, and health are not a robust indicator of whether the adopted fiscal policy mix adequately protects the poor. And “defining spending floors to narrowly cover the needs of the most vulnerable,” one of the key recommendations of the Staff Paper, while important, is not enough. Why? In simple terms, because the effect of these aggregate floors on the near-poor and the poor cannot be determined without information on not only how much the poor receive in benefits but also how much they pay in taxes.

The key to knowing how countries can minimize the tradeoff between appropriate fiscal consolidation and protection of the poor requires having and analyzing, country by country, micro data on the incidence of expenditures and taxes as well, on the poor and other income groups. A new generation of such studies undertaken in more than 30 countries (under the sponsorship of this program, and discussed here shows that significant numbers of the poor, especially in low-income countries, are net payers into the fiscal system (primarily due to consumption taxes) and that reducing subsidies can hurt the poor even when the subsidies, for example on gasoline, benefit disproportionately the middle class and the rich.

In some countries fiscal policies made the poor poorer

For example, in Ethiopia, Ghana, Guatemala, Nicaragua, Uganda, Togo, and Tanzania (in around 2012) the extreme poverty rate was higher after taxes and transfers than before. In Ethiopia, half a million people were pushed below the extreme poverty line because taxes they paid outweighed the value of transfers and subsidies they received. In Tanzania, 50 percent of the extreme poor were impoverished by the fisc. And in Ghana, Guatemala, Nicaragua, Tanzania, and Uganda, the extreme poor—on average—were net payers into the fisc—in all cases despite cash transfers. While it is true that the extreme poor in these countries may benefit from government spending on education, health, and infrastructure, these cannot replace food, clothing, and shelter.

With this kind of micro evidence, governments are equipped with the information to justify “fiscal policy fixes” that benefit the poor (or at least protect them) while preserving macroeconomic stability. What happened in Ethiopia illustrates this well: “The results from the Ethiopia CEQ Assessment contributed to two key policy changes in 2016: the cash transfer program was expanded to include urban areas and the minimum threshold of taxable personal income was raised.”

What’s next?

The good news is that the IMF, in partnership with the Commitment to Equity Institute at Tulane University and the World Bank, has recognized the need for more granular distributional analysis incorporating fiscal incidence analysis in its surveillance and fiscal consolidation programs in Costa Rica, Guatemala, NamibiaSwaziland, Togo, and Zambia (and with more countries in the pipeline). The case of Togo is a good example: “[t]he finding that VAT [Value Added Tax] imposes an economic burden on lower-income households, despite its overall progressivity, also suggests the need to effectively target social spending.” (p. 13).

Better late than never: The IMF is waking up on how to help countries manage fiscal reforms AND protect the poor. And a bottom line: spending on safety nets for the poor needs to be set net of the taxes that the poor pay. 


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.