The IMF's focus on debt recovery must be corrected

In this guest post Peter Doyle, a former senior IMF economist, argues that the Fund's fiscal conditionality costs more than meets the eye.

My recent alarum about the decade of IMF programmes in Jamaica focused on the need for macroeconomists to organise and defend ourselves against outlier practitioners acting in our name, in the same way that the medical profession organises against medical malpractice.

But there are also key lessons to draw concerning sovereign insolvency arrangements.

The IMF rebuttal's to the charge of macroeconomic malpractice in Jamaica is that debt-to-GDP ratios have fallen there this decade from 140 to 110 per cent of GDP under its guidance. And with output stagnant rather than collapsing Greece-style — not so bad!

That defence is revealing. In particular, consider why IMF-backed primary surplus targets this decade of 5.5 per cent in Greece and 7-9 per cent of GDP in Jamaica led to so much worse output outcomes in Greece.

Partly, that difference reflects Greece's deeper adjustment and structural weaknesses, and its fixed exchange rate. But, critically, Jamaica's growth potential was far higher: what Jamaica lost as a result of IMF fiscal conditionality was not output, but output forgone.

To gauge the order of magnitude of that loss, compare Jamaica under IMF-tutelage with its macroeconomic (not geographic) peers this decade (in the same adverse global context). And not "average performing" but "best in class" peers, because the standard to which IMF conditionality over so long a period should be held is not "average" but "best in class".

Whereas Jamaica grew some 1 per cent annually this decade, its African best-in-class peers, both those with GDP per capita below Jamaica — Uganda, Kenya, Ethiopia — and with GDP per capita above Jamaica — Botswana, Namibia, Mauritius — grew on average 5 per cent or more annually. That is an annual gap with Jamaica of some 4 percentage points.

Over this decade, that 4-point gap represents a cumulative loss of over 40 per cent of Jamaican start-of-period GDP, an order of magnitude very much in epic Greek territory. Sensible technical refinements to that calculation do not alter the substantive conclusion.

With Jamaica observing all IMF conditionality for a decade, the stagnation and output foregone points directly at the culprit: not so much the usual mis-measured fiscal multipliers, but rather the cumulative squeeze on the police, hurricane defences, education, health, investment and taxation due to the IMF's punishing primary surplus targets.

That fiscal prescription not only squandered the all-too-precious national consensus that "something had to be done", but also handed all of its fruits over to the creditors.

It obviously makes no sense for the IMF to grandstand on the need for deep debt write-offs for Greece on grounds that countries cannot sustain elevated primary surpluses and yet go on insisting on exactly that and more in Jamaica and elsewhere.

But the key point to which I want to draw your attention here concerns the implications of Jamaica for the IMF's role in sovereign insolvency, notably its approach to Debt Sustainability Analysis (DSA) as applied in all of its programs.

That practice simply asks if a series of primary surplus targets and associated output projections exists which can then be plugged into country spreadsheets to yield declining debt ratios, all so that the IMF can persuade itself on a traffic-light scale — green (no problem), yellow (with difficulty) and red (no way) — that the country will deliver under threat of withdrawal of financing. This was the procedure for Jamaica in 2009 and Greece in 2010.

But that procedure only asks "do debt ratios decline?" It never asks "how much output is foregone due to the primary surplus requirements underlying that decline in debt ratios?"

That oversight bites with particular venom in emerging countries given their high growth potential. As a matter of simple algebra, a country with trend-potential growth of say 10 per cent can give up much more in potential growth under a fiscal squeeze before the level of GDP — the denominator in the debt ratio — drops than can a country with trend-growth potential of only 1 per cent. So the same sustainability rule — requiring a declining debt ratio — inevitably costs considerably more in output foregone in the former case.

So what, you may protest: they borrowed, they should pay. And if that means they give up more in output foregone, well they should have thought of that before!

But such dismissal of output forgone is not granted in corporate or personal insolvencies. Limited liability shields investors, no matter their net worth, and creditors' claims on individuals' future incomes are capped. These are not acts of mercy: otherwise, output foregone would be too high. Few would invest if their entire net worth was on the line and if all of an individual's future income was garnished, few would run the risk of borrowing.

So why is this concern with output forgone that caps creditor recoveries in corporate and personal bankruptcy arrangements ignored by the IMF in sovereign insolvencies?

Well, apart from the fact that the IMF never assesses output forgone that is due to its own conditionality — so it doesn't even know how much it is — the answer, as illustrated by Jamaica, is that even if primary surplus targets there had been set at, say, 1.5 per cent of GDP from 2009, allowing growth to rise to best-in-class 5 per cent, debt ratios would still have risen. And rather than press for deep debt write-offs (as opposed to yet another round of Jamaica's papering-over-the-cracks debt reschedulings) in order to reconcile growth potential with debt sustainability, the IMF chose to act as debt collector instead. Thus, it imposed exorbitant potential-growth-destroying primary surplus requirements.

That is a choice. Given the gross inadequacies of current sovereign resolution mechanisms, particularly relative to their personal and corporate counterparts, deep sovereign debt write-offs are a heavy lift. But with the IMF assessing 40 per cent of developing countries to be in or near debt distress, one wonders how much of the longstanding puzzle that poor countries aren't catching up — the so-called β-coefficient of unconditional convergence — is due to that choice or to such policies adopted by governments "voluntarily" to avoid such IMF programmes.

And weighing output foregone is not just "nice" or even "efficient". Right now, the US government is shut down over how to curb immigrants from the South, and the EU has similarly been confronted by heavy loss of life among people making desperate passages across the Mediterranean. Many factors prompt these migrations, and output foregone is one of them.

What should be done? For every IMF programme, and for every case where the IMF calls for primary surpluses above 1.5 per cent of GDP under its current DSAs, the IMF should be required to produce an accompanying annex showing the order-of-magnitude output foregone due to its primary surplus targets. That would at least bring the loss to light.

In addition, the IMF red-zone, in which it cannot lend without write-offs, should be redefined to include output forgone. If that loss is high, the IMF should require debt write-offs before lending even if, by its current DSA analysis and as in Jamaica, debt ratios can be put on a downward path. This red zone redefinition would rebalance away from debt collecting towards macroeconomic concern with output foregone.

Such output assessments are not "too difficult or speculative"; the best-in-class peer comparator analysis above is evidently tractable, and will certainly catch the most egregious cases even given IMF's systematic over-optimism in programming. They can also be executed on a forward-looking basis using projections for peers taken from World Economic Outlook reports.

Many countries would thereby be shifted from IMF "debt-amber" to deeply into "debt-red" territory. Jamaica would top that shifted list. But among others, Jamaica's neighbour, Barbados, would also move, as would Zimbabwe and even Argentina. "IMF stigma" would be much diminished were it to do this.

But, you may think, isn't this all jumping the gun? Even if enormous growth potential is destroyed by current IMF sovereign insolvency practices, that doesn't mean debt write-offs would realise that potential. Isn't elevated public debt itself a sufficient statistic of systematic governance failures by borrowers, which write-offs would only validate?

That generic censorious condescension sounds so easily compelling, doesn't it.

Think again. Of Jamaica's public debt of 140 per cent of GDP in 2009, one quarter was from a bail out of its financial sector in the mid-1990s. That followed sweeping financial liberalisation that was required of Jamaica under its 1991 IMF program. If anything, that part of public debt reflects systematic IMF governance failures, not those of Jamaica.

Worse still, though both Jamaica and the US are housed in the same IMF area department, the IMF failed to draw lessons from its 1990s debacle in Jamaica for its subsequent advice on US financial liberalisation. An arrogant assumption that there was nothing to learn from somewhere so small (or black?), perhaps. But in yet another instance of a dot-not-connected, the Global Financial Crisis followed, also piling on to Jamaican public debt, big time.

And creditors have recently demonstrated in the global crisis that their legal, intellectual, financial, and political powers can hold even the advanced world, blessed with the best governance arrangements known to man, to ransom. Against that lot, and even with the best will in the world, emerging countries don't stand a chance.

The fundamental solution is to curb creditor rights in sovereign insolvencies.

There are many steps to do that, including a revamp of international sovereign resolution arrangements to shift the balance towards debtors, building on the IMF's own tentative lending into arrears initiatives. Greater awareness of just how costly current sovereign resolution arrangements and IMF DSAs are in terms of output forgone might help to motivate this. Jamaica is uniquely placed to reveal this precisely because it complied with all IMF — including outrageous primary surplus — conditionality for so long. And blocking the IMF from simply defaulting to debt-collector mode by requiring it to produce best-in-class peer "output foregone" annexes, and incorporating macroeconomic concern with output foregone into the definition of its lending "red zone", would mark a good start.