All these monetary uncertainties mean that the IMF staff set program targets for central bank credit, foreign exchange reserves, and money supply based on shaky numbers. The money supply number is very important because it determines how much credit expansion is safe without losing dollar reserves or increasing inflation.
GDP growth plays a big role in projecting how much money demand or other important variables will grow. In March 2003, IMF staff put Mali’s GDP growth in 2001 at 1.5 percent. By August 2003, it had raised the 2001 number to 3.5 percent. Just five months later, in January 2004, IMF staff now put Malian growth in 2001 as 13.3 percent! This is not to say the IMF is incompetent at statistics; it is just that any statistics are very shaky in very poor countries.
Things get worse on the arithmetic that says the government budget deficit must equal its sources of financing. Government spending minus revenue yields one estimate of the budget deficit. Adding up all the sources of financing the deficit (central bank credit, foreign borrowing, etc.) yields another. The numbers do not agree. IMF programs will include an “adjustment” to reconcile the two. In the official statistics from the IMF’s Government Finance Statistics, the “adjustment” is equal to 55 percent of domestic credit on average.12 So we are not sure exactly how much the government deficit or domestic credit is now, and thus are not sure how much the government has to cut it.
Just as the IMF should not be blamed for all budgetary austerity, an IMF program also is not necessary for the budget to be balanced. If the IMF were not there, the government would still be constrained by its available revenues and whatever loans people were willing to give it. If it was an irresponsible spender, private lenders would not lend to it. The government could print money, but the revenues from doing so would be limited and come at the price of a high inflation rate, which is usually unpopular. The government on its own doesn’t have the same dependence on shaky statistics that the IMF does because it will be constrained by real resources, even if it doesn’t know how much it has. You may not know the balance in your checking account due to sloppy bookkeeping, but if you spend too much, your checks will bounce. Is the IMF’s shaky accounting an improvement on what the government would do without the IMF?
Unstable Behavior
The second problem is that planning the macroeconomic program depends not just on accounting but also on the behavior of people inside and outside the economy. Remember, for example, that the effect of expanding central bank credit to the government and printing money would cause people to turn the extra money in to the central bank for dollars, depleting international reserves. But what if people decided they wanted to hold more currency, for whatever reason? The IMF estimates the desire to hold currency by assuming that the ratio of currency to gross domestic product will remain stable. Un-fortunately, after examining data on all IMF borrowers for all available years, I found the historical path of this ratio to be more like that of a drunken unicyclist.
A supply of money greater than the demand to hold that money could also drive up prices. Again, it is uncertain how the actual supply of money compares with what people willingly hold. This may be why the IMF has had difficulty predicting inflation under its programs of financial discipline and restructuring. Post-program inflation under the IMF was higher than the program targets on average in the 1990s for a worldwide sample of countries.13
Conversely, what if people holding domestic currency suddenly panicked and wanted to turn it in for the central bank’s dollars? It’s not always clear why they panic, but it happens. International reserves would drop precipitously for reasons unrelated to government budget deficits. Many economists think that this is a good description of the East Asian financial crisis of 1997–1998. East Asian countries were not running large government deficits, yet they suffered currency panics and disappearing foreign exchange reserves all the same.
Another loophole in the relationship between budget deficits and foreign exchange reserves is that the government finances its deficit not only with central bank credit but also with foreign debt. The willingness of foreign investors and banks to buy government bonds is another unknown. This is not so relevant for the poorest countries, or with countries that just have a bad rep as politically unstable or as profligate spenders—they don’t qualify as “emerging markets,” to use Wall Street jargon. But other poor countries do qualify as emerging markets; private investors and banks help finance these countries’ government deficits by buying government bonds. One Web site suggests that there were about forty-five emerging markets (countries), together accounting for 2.6 billion people.14 A sudden surge in demand for government bonds by foreign investors could allow the governments of these countries to cut back their use of central bank credit, building up dollar reserves without any need for fiscal austerity. Conversely, a flight out of government bonds in emerging markets—as happened after the Mexico crisis in 1994, the East Asian crisis in 1997–98, the Russia crisis in 1998, and the Argentina crisis in 2001—could suddenly force governments to use central bank credit again, running down foreign exchange reserves.
How much do you need to cut central bank credit when the desire to hold domestic currency jumps around like my dog, Millie, after swallowing a jalapeño? How much do you need to cut deficits when the willingness of foreigners to hold government bonds is oscillating? The desire to hold currency or government debt may itself depend on the government’s policy on cutting central bank credit or budget deficits. It is circular—if the government is willing to cut the deficit and cut central bank borrowing, then people will be willing to hold more currency and hold more government debt, which lessens the need to cut deficits and central bank borrowing. Finding out where this complex process reaches equilibrium is more complicated than jet-lagged IMF staff can capture by adding up numbers on a spreadsheet.
The moral of the story is that IMF prescriptions about how much to cut central bank credit and government deficits are often based on shaky foundations. The IMF should give up its pretensions that it understands the whole complicated system of financial equilibrium—this is another strain of the disease of utopian social engineering.
Is the IMF a Wimp?
The real test of the IMF’s approach is whether it gets results on stabilizing macroeconomic disorder. On average, one of the big surprises is that the IMF has been weak in enforcing its conditions on macroeconomic misbehavior.
Let’s analyze the IMF and the World Bank adjustment loans together because they play the same role of promoting “structural adjustment” (i.e., the reforms to straighten out finances and promote free markets) and because World Bank adjustment loans often support financing for IMF programs. The key number is what happens to the budget deficit. Remarkably, budget deficits did not improve from one adjustment loan to the next over 1980–99.15
Next, let’s broaden the definition of bad government policy to include a variety of indicators: (1) whether the inflation rate is above 40 percent; (2) whether the dollar is trading on the black market for foreign exchange at more than a 40 percent premium over the official rate; (3) whether the official exchange rate is more than 40 percent out of line with the competitive rate that facilitates exports; and (4) whether interest rates are controlled at more than 5 percent below the rate of inflation. If any of these conditions is met, economic policy is classified as bad. These are exactly the kind of bad economic policies targeted by the IMF and the World Bank. That is, the IMF and the World Bank give “structural adjustment loans” on the condition that all of these problems be corrected. Yet the fraction of structural adjustment loan recipients violating one or more of these conditions did not decrease from one structural adjustment loan to the next (see figure 24).
Fig. 24. Fraction of Countries with Macroeconomic Distortions by Cumulative Number of Adjustment Loans
What explains this surprise? One possible explanation is the IMF’s tendency to wipe the slate clean with each new loan, especially if new officials are in power in the recipient country. Eve
n though an IMF loan is supposed to be a short-term or medium-term bailout, the countries often don’t seem to stay bailed out. Other countries fail to fulfill the conditions on old loans and yet get new loans anyway. Countries such as Ecuador and Pakistan went for more than two decades receiving one IMF loan after another, even though they had never completed any previous IMF program (meaning they didn’t fulfill the conditions to get a second or later installment of a loan commitment).16 A slew of examples in Africa had the same problem, which was to contribute to the African debt crisis.
The IMF’s relationship with its clients is capricious. First, the IMF is tough on cutting budget deficits and causes riots. Then a new government comes in and again runs a high deficit, which the IMF then tries to bring down again (the deficit, not the government).
Debt and Consequences
The IMF also monitors government debt because it knows that a heavier debt load will increase future government deficits by increasing interest costs on the debt. Too high a debt will also make the creditors unwilling to continue lending.
The IMF has its own self-interest in this—if the country owes too much, it can’t pay back the IMF. The IMF protects itself against this with a stipulation in its loan agreements that the government always pays it first, before other creditors. Still the IMF fails in its mission if a country becomes insolvent. Part of the IMF’s role is to head off this denouement by persuading the government to borrow no more than a reasonable amount. But what is a reasonable amount? The IMF has a difficult time finding the answer to this question.
Then there is the problem of repaying the IMF loans themselves. The capricious relationship just described between the IMF and some of its clients means that first installments of IMF loans are made, but true macro-economic adjustment does not happen. This does not augur well for countries being able to repay the IMF loans.
Bailing Yourself Out
One way the IMF has adjusted to this situation is to keep making new loans to repay the old loans. Once a country is in deep to the IMF, with the country owing the Fund due to previous bailout packages, it is hard to get out.
Although the IMF bailing out Wall Street investors is controversial, the real problem is that it is bailing itself out. If the IMF did not make a new loan, the country might not repay the previous IMF loan. The IMF often drags in its sidekick, the World Bank, just across Nineteenth Street in Washington from the IMF, which makes an “adjustment loan” to the country as part of the bailout package.
One sign of this self-repayment is the high loan repetition rate for the Sisters of Nineteenth Street. The probability of getting a new loan does not go down with the number of IMF and World Bank adjustment loans already received. (See figure 25.)
The IMF’s Independent Evaluation Office highlighted this problem of “prolonged use” of IMF money.17 It defined a “prolonged user” as a country that was under an IMF program for seven years out of any ten-year period. Forty-four countries met this definition of IMF addiction over the period 1971–2000. Prolonged use has become more common in recent years. In 2001, loans to prolonged users accounted for half of all IMF lending.
As usually happens with addiction, the IMF habit of repeat lending includes some self-deception. The IMF made overly optimistic projections of the prolonged users’ GDP and/or export growth. It granted waivers of its conditions on its loans to repeat offenders. As the Independent Evaluation Office describes this process of fooling yourself into lending: “Internal incentives in the IMF encourage overpromising in programs. This results from both the relatively short time frame of programs, forcing optimistic assumptions about the pace of adjustment…. This led to a tendency to downplayrisks. Even when, as was often the case, they were well identified during the internal review process, the assessment of risks was not candidly presented to the Executive Board.” The Evaluation Office noted that repeat lending by the IMF tended to weaken the leverage the Fund had over countries to enforce its conditions.
Fig. 25. Repetition Rates of Structural Adjustment Lending After Given Number of Loans, 1980–1999 Number of cumulative adjustment loans
The IMF displays one of the classic symptoms of Planner’s disease: in many countries, it keeps doing the same thing over and over again to reach a never-reached objective. The repetition itself shows the failure of previous attempts at “short-term stabilization.”
Heavily Indebted Poor Country Crisis
Repeated lending also does nothing to make the debt repayable, as debt keeps mounting without countries becoming more able to service the loans. One embarrassment happened with the poorest countries, which received many IMF loans as well as World Bank “structural adjustment loans,” also meant to restore financial discipline. The poorest countries also received loans from Western governments and export credit agencies. Their debt load became so extreme that after 1996, the IMF and the World Bank, for the first time in their history, forgave part of their own loans. The IMF and the World Bank called these impoverished borrowers the Heavily Indebted Poor Countries (HIPCs). Among poor countries receiving above-average amounts of IMF and World Bank structural adjustment loans, seventeen of the eighteen became HIPCs, whose debt the IMF and the World Bank partly forgave. Among poor countries who had less than the average IMF and World Bank borrowing (measured by number of loans), only eight of seventeen became HIPCs. This could reflect the maddening selection problem again—that sick economies were both more likely to pile up debt and to turn to the World Bank and the IMF for help. However, this does not reflect wise lending by the IFIs, since a good part of the HIPC debt that had to be forgiven was directly from the IMF and the World Bank—even in such “successful” cases as Ghana and Uganda. Far from helping the poor countries achieve a reasonable debt load, the IMF and the World Bank were themselves contributing to the excessive debt of the HIPCs.
HIPCs qualified for debt relief by meeting some of the same conditions that they had failed to meet (or first met and then backtracked on) when getting the original loans. As of March 2005, debt-reduction packages have been approved for twenty-seven nations, providing fifty-four billion dollars’ worth of debt relief—a reduction by about two thirds of their outstanding debt.18
HIPC debt forgiveness was supposed to be a once-and-for-all solution that would solve the debt problem. The IMF and the World Bank often had optimistic forecasts for GDP growth in the HIPCs. This hoped-for growth would have allowed the HIPCs to keep the ratio of debt to GDP from surging again. But the debt relief did not spur growth.
Bolivia is an example. The country had been an IMF ward ever since the first HIPC relief in 1998. The IMF and the World Bank projected rapid growth in Bolivia in per capita income over 1999–2003; instead living standards declined (see figure 26).
The failure of debt relief to spur growth was a problem because the failure of the original loans to spur growth was what had caused the debt problem in the first place. We have already seen that Africa, a favorite destination of repeated IMF and World Bank lending, also failed to have the growth that would have enabled it to service its debt. This is a general pattern: the growth in program countries fell short of the IMF’s own targets. On average for IMF programs in the 1990s, the target GDP growth was 4 percent, but actual growth was only 2 percent.19 Since population growth was also about 2 percent, this meant that the actual growth of income per person was close to zero.
Fig. 26. Bolivia: Index of Per Capita Income Projected by IMF and Actuals
The supposed once-and-for-all debt relief in 1996 was superseded by Enhanced HIPC in 1999, which gave deeper debt relief to more nations. However, even Enhanced HIPC was not enough. The G8 Summit in July 2005 decided to give 100 percent debt cancellation (worth forty billion dollars) to eighteen low-income countries that had already qualified for HIPC debt relief, including Bolivia and fourteen African countries.
Low-income countries have had debt problems since the 1960s, yet still the IMF and the World Bank have insisted on making repeated loans to very shaky borrow
ers.20 This is yet another of our numerous examples of the aid community pouring in resources at a fixed objective—financing “development” with aid loans. The IMF and the World Bank kept making new loans to repay old loans, even though countries were having ever-increasing difficulties at repayment.
At this point, the ever-escalating degree of debt forgiveness has destroyed low-income debt as a believable instrument to finance anything. The borrowers have little incentive to repay when they see the debts periodically forgiven (what economists call “moral hazard”). Calling a loan to the poorest countries a “loan” has become ever more fictional. The World Bank, which is an aid agency, should just give the poorest countries grants, not loans (this was one of the better ideas of the Bush administration on foreign aid). The IMF, which is not supposed to be an aid agency, should get out of the business of loaning money to the poorest, least creditworthy countries altogether. Is there any reason to keep bailing out countries that chronically fail to stay bailed out?
Cry, Argentina
Another of the IMF’s recent embarrassments was Argentina’s default on its government debt in December 2001. The IMF was deeply involved in the elusive quest for Argentine financial stability, with fifteen standbys from 1958 to 1999. After unhappy decades of financial chaos, Argentina had a decade of financial stability after 1991. It was the star pupil of the IMF, even as other IMF clients such as Mexico, Russia, Brazil, and the East Asian countries experienced crises. But Argentina began to get into trouble in 1999. President Carlos Menem, who presided over Argentina’s near decade of financial stability, increased public spending in his quest to get a third term in office. When this quest failed, electoral politics among other contenders took over. In the understated language of a former top IMF official, Michael Mussa, “election-year concerns further depressed the normally low level of interest Argentine politicians…attached to measures of fiscal prudence.21 Argentina borrowed heavily from emerging-market investors in 1999 and 2000. By this point, the game was up—Argentina would not honor its commitments to private foreign lenders. In the quaint language of emerging markets, private lenders would “take a haircut.” At this point, the IMF should have pulled the plug. Instead, it put together a forty-billion-dollar rescue package that included fourteen billion dollars from the IMF itself, five billion dollars from the Inter-American Development Bank and World Bank, one billion dollars from Spain, and a projected twenty billion dollars from private lenders—announced on January 12, 2001.
B000QJLQXU EBOK Page 24