Macroeconomic Conditionality

3.1 Monetary policy

There were two controversial components to the IMF's macroeconomic programs in Asia: high interest rate policy and fiscal tightening (Furman and Stiglitz 1998). The debate on high interest rate policy remains and is likely to remain unsettled. The interest rate defense of a falling currency has been a standard practice in many contexts, and has been successful in some cases, though not in others. In fact, there is some empirical evidence to show that high interest rate policy in the Republic of Korea had a stabilizing effect on the exchange rate (Cho and West 2000; Chung and Kim 2002). In a model of the war of attrition between speculators and monetary authorities, Gregori (2009) shows that the success or failure of an interest rate defense depends in a complex fashion on the interest rate, the associated private costs to both parties, and the rate of expected depreciation by the speculators when the defense fails. At any rate, in Asia over 10 years ago, no one would have advocated reducing interest rates when the exchange rate was falling drastically.

The debate in Asia can be summarized in terms of the following dilemma: high interest rate policy further damages the banking sector already weakened by the crisis, but unless interest rates are raised, a weaker currency would magnify the banking crisis in any event through the adverse balance sheet effect. In Asia, there was also a perception that the IMF had caused the unwilling national authorities to raise interest rates, a policy which would subsequently magnify the banking crisis and as a result prolong the economic stagnation.¹⁴ The lesson of Asia does not necessarily concern the content of monetary policy conditionality itself. Rather, it is that the monetary authorities should pay sufficient attention to the potential negative impact of high interest rate policy on a weak banking sector and it is only the national authorities who can take the ultimate responsibility for the choice of high interest rate policy in view of its positive and negative consequences.¹⁵

In Europe, tight monetary policy was also the core element of all IMF programs. In Iceland, the policy interest rate was raised by 6 percentage points (to 18%) as a prior action for the program's Executive Board approval, and performance criteria were set on the provision of central bank credit to the government and the private sector. In Ukraine, although the stance of monetary policy had been tight prior to the crisis, the authorities in the fall of 2008 responded to the global tightening of liquidity by lowering the policy interest rate by 3.5%. What the IMF program did was to require that monetary policy be reversed back to the tightening stance of the pre-crisis period. In terms of the stance of monetary policy, the European programs differed little from the Asian programs over 10 years ago.

If there is any difference, it is about procedure. First, the documents for the European programs emphasized that high interest rate policy was the national authorities' preferred policy. For example, the Hungarian authorities had already raised the policy interest rate in early 2008, and the IMF program simply maintained the stance of policy already chosen. In Iceland, the authorities had been raising the policy interest rate (before the reversal in mid-October) and pledged to reform the Housing Financing Fund (as a way of containing the provision of credit). Here again, the IMF program simply held on to the previously chosen course of policy.

Second, the documents for the European programs specified the purpose and logic of high interest rate policy. In most cases, they stated that the purpose was to stabilize the exchange rate while noting the negative consequence of premature monetary easing for this purpose. At the same time, the documents also stated the need for banking sector restructuring as an objective of the program and, insofar as high interest rate policy conflicted with this objective, the need to manage monetary policy flexibly. In Ukraine, for example, the document stated the need for further monetary tightening but only after the pressing liquidity problem of the banking sector was resolved.

3.2 Fiscal policy

The more controversial aspect of IMF conditionality in Asia concerned the fiscal component. There is now broad agreement that fiscal tightening as initially programmed in Asia was unwarranted not only in view of a prospective deceleration of output, which did in fact materialize, but also because, as Ito (2007) and others have argued, fiscal balances were initially in surplus and fiscal prodigality was never a cause of the crisis. The programmed fiscal adjustments were significant. In Thailand, the adjustment amounted to 2.8% of GDP for October 1997–September 1998; in Indonesia, it was 1% and 2% of GDP, respectively, for 1997–1998 and 1998–1999. In the Republic of Korea, fiscal measures amounting to 1.5% of GDP were programmed for 1998. As a result, the programs for these countries envisaged a small fiscal surplus (0.2–1.0% of GDP) in the subsequent calendar or fiscal year (Table 3 [ PDF 18.8KB | 1 page ]).¹⁶

In Europe also, all program documents stressed fiscal tightening as the critical element of the program. In Hungary, for example, the documents explicitly stated that the objective was to implement “a substantial fiscal adjustment” and programmed a tightening that amounted to about 1% (2.5% in cyclically adjusted terms) of GDP from 2008 to 2009. The tax cut planned for 2009 was also scrapped. The government of Hungary had failed several times to raise funds in the market. A government that cannot borrow in the market cannot be expected to continue to run fiscal deficits. Surely, restoration of fiscal sustainability needed to be a primary objective of any crisis management program.

In Ukraine also, the program envisioned a fiscal tightening amounting to 1% of GDP. Ukraine had continued to run fiscal deficits amid strong capital inflows, and the need to tighten fiscal policy had been recognized as a priority task even before the onset of the crisis. In Latvia, fiscal tightening was essential, not only to maintain the peg by depreciating the real exchange rate, but also to achieve the national goal of joining the Euro Zone. The program documents stated that, with determined efforts, the Maastricht criteria for fiscal policy could be met by 2012. Under the IMF program, an adjustment equivalent to 7% of GDP was carried out for the 2009 budget (compared to the original budget).

Although all programs stressed fiscal tightening and placed performance criteria on fiscal deficits, however, the policy actually programmed was accommodative.¹⁷ For instance, the programs allowed fiscal deficits to continue in 2009 except in Ukraine and, for Iceland and Latvia, permitted the deficits to increase from 2008 to 2009. This is probably a reflection of realism, not of the fiscal policy stance. Iceland, in particular, had experienced a virtual national bankruptcy, with the cost of depositor protection and capital injection into the banking sector estimated at 80% of GDP. With the expected recession-induced increase in fiscal deficits, gross government debt was expected to increase from 29% of GDP at the end of 2007 to 109% at the end of 2009. Programming a fiscal surplus under these circumstances would have only undermined the credibility of the program. In the event, a further deterioration of the global economy would cause the fiscal deficit targets to be revised upward for 2009, except in Iceland.

In short, the logic of fiscal conditionality in the European programs was that it would realistically allow deficits to remain or increase in the short run, but place fiscal consolidation as the central objective. The documents for Iceland explicitly stated that medium-term fiscal consolidation would formally commence in 2010. As a consequence of realism, the programs were flexible. The document for Ukraine acknowledged the possibility of relaxing the fiscal target for 2009 depending on the prevailing circumstances (which in fact did take place). In all four countries, the costs of bank restructuring (including depositor protection) were excluded from fiscal conditionality (whereas in all three Asian programs additional tightening was programmed to contain such costs). This would allow the countries to increase fiscal spending flexibly without violating the terms of conditionality.¹⁸

3.3 Underlying macroeconomic assumptions

In order to understand the logic of fiscal conditionality, it is necessary to review the underlying macroeconomic assumptions. The primary reason for IMF conditionality in Asia to include fiscal tightening was that the IMF did not fully appreciate the nature of a capital account crisis, with attendant sharp capital flow reversals, and the negative balance sheet effect of exchange rate depreciation. The capital flow reversals were far greater than anybody had supposed. The magnitude of the reversal from 1997 to 1998 was especially large in the Republic of Korea (12% of GDP) and Thailand (13%). Instead, the IMF staff underestimated the capital flow reversals the three countries would experience and assumed them to be only 0.5%–2.5% of GDP (Table 4 [ PDF 18.6KB | 1 page ]).

A large capital flow reversal could lead to a contraction of output because the current account needs to swing into surplus as a counterpart of the capital account deficit. Output contraction would be the surest way to compress imports and thereby generate the required current account surplus. The IMF underestimated the severe contractionary effect coming from the capital account and therefore failed to anticipate a possible negative output growth. It continued to believe that the Asian crisis countries would register positive output growth (Table 5 [ PDF 17.2KB | 1 page ]).

Another reason for underestimating the contractionary impact of the crisis was the failure to understand the balance sheet effect of currency depreciation. As noted earlier, many at the time believed that devaluation was expansionary because it stimulates exports and raises the price level. According to Boorman et al. (2000), the IMF staff had initially expected the countries' exports to expand much more as the currencies fell in value. The sharp contraction of output in the crisis countries therefore came as a surprise. Part of this was the negative impact of increased uncertainty and reduced confidence on corporate investment, but there was also the negative balance sheet effect of currency depreciation, given the history of significant unhedged foreign currency borrowing.

In contrast, the European programs clearly recognized the contractionary impact of a capital flow reversal and exchange rate depreciation. Except in Iceland, all programs assumed net capital inflows for 2009 and give the impression that they were on the optimistic side. This may have reflected the assessment that there was sufficient external financing (both official and private through PSI) to keep the inflow of capital. Even so, the European programs envisioned a far greater reversal of capital flows (4%–12% of GDP) than initially assumed in the Asian programs (0.5%–2.5%).

As to the balance sheet effect, the IMF staff had acquired a tool (the “balance sheet approach”) to analyze the implications of currency and maturity mismatches in the balance sheets of various sectors in an economy. In fact, the balance sheet approach may well be the single most important analytical development of the past decade within the IMF (see, for example, Allen et al. 2002; Rosenberg et al. 2005). As a result, all program documents in Europe acknowledged the negative wealth effect of the exchange rate depreciation observed up to that time.

In Iceland, where the currency had depreciated by 70% against the US dollar (and 50% against the euro), serious recession was forecast for 2009–2010. In Latvia, the justification for maintaining the peg included the argument that, given the large balance of foreign currency debt (amounting to 70% of GDP) and extensive dollarization in domestic financial transactions, the balance sheet effect of allowing the currency to depreciate would be simply too large. The documents then argued that, given the decision to maintain the peg, deflationary pressure would emerge, making economic recovery a difficult and prolonged process. In all countries, the programs saw negative economic growth for 2009.

Download this Paper [ PDF 323.3KB| 24 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.