Fault Lines: How Hidden Fractures Still Threaten the World Economy (13 page)

Developing countries had to borrow from abroad to finance the difference between what they spent (their consumption plus their investment) and what they produced, as well as to pay interest and principal on prior borrowings. In the 1950s and 1960s, much of this borrowing came from other countries or from multilateral institutions like the World Bank.
²
However, in the 1970s and 1980s, Western banks, recycling the mounting petrodollar surpluses of Middle Eastern countries, assumed more of the lending to developing countries. In the 1990s, foreign arm’s-length investors such as mutual funds and pension funds increased their share of lending to developing countries by buying their government and corporate bonds. Thus foreign financing of developing countries became increasingly private and arm’s-length.

Unfortunately, few countries have the discipline to borrow and spend carefully while running large trade deficits. Indeed, large amounts of foreign financing tend to encourage wasteful spending decisions. Many developing countries learned from terrible crises in the 1980s and 1990s that it was very risky to expand domestic spending rapidly through a foreign-debt-financed binge, whether the expansion was through consumption or investment.

The boom in busts in the 1990s had varied effects on the behavior of developing countries.
³
For those like Brazil and India, which were consuming too much, the financial difficulties made them liberalize their economies and cut back on excessive and populist government spending, leading to more stable and faster growth. The busts led exporters, like the East Asian economies that had been borrowing to fund investment, to curtail investment so as to reduce their dependence on foreign borrowing. They started intervening in their exchange rates, building up exchange reserves, and in the process pumping their savings out into the global economy, ready to finance anyone who wanted to spend more. With the East Asian economies absorbing fewer imports, the surpluses searching for markets elsewhere increased, making the global economy yet more imbalanced. The fault line described in the previous chapter deepened.

In the perfect world envisioned by economists, a country’s investments should not depend on its savings. After all, countries should be able to borrow as much as they need from international financial markets if their investment opportunities are good, and their own domestic savings should be irrelevant. So there should be a low correlation between a country’s investment and its savings. In a seminal paper in 1980, Martin Feldstein from Harvard University and Charles Horioka from Osaka University showed that this assumption was incorrect: there was a much higher positive correlation between a country’s investment and its savings than one might expect if capital flowed freely across countries.
⁴

The interpretation of these findings was that countries, especially poor ones like Burundi and Ecuador, could not get as much foreign financing as they needed, so they had to cut their coats to fit the cloth. However, there is another explanation, not necessarily incompatible with the first. Countries might also have chosen to limit their foreign borrowing, thus inducing a strong correspondence between their investment and their savings. But why would they do so if their investment opportunities, once they overcame the organizational deficiencies noted in the previous chapter, were high?

To try to understand this question, some years ago I undertook a study with Arvind Subramanian, now at the Peterson Institute, and Eswar Prasad, now at Cornell University, in which we looked at the correlation between the average current-account surpluses of developing countries and their growth over recent decades. The current account is just the difference between a country’s savings and its investment. A surplus indicates that the country contributes savings to the global pool, while a deficit indicates that it borrows from the rest of the world to finance its investment. A current-account surplus typically also means a trade surplus: the country exports more than it imports.

We found a positive correlation for developing countries: the more a country finances its investment through its own domestic savings, the faster it grows. Conversely, the more foreign financing it uses, the more slowly it grows. Of course, a country might need foreign financing either because it saves very little relative to the norm or because it invests a lot. We found that the more a country invests, the more it grows, which is natural: by investing, it increases its roads and machines, all of which go to make its workers more productive. However, the more its investment was financed from foreign sources as opposed to domestic savings, the slower its growth. Interestingly, these relationships did not hold for developed countries. Developing countries, at least the ones that grew fast on a sustained basis, seemed to avoid significant foreign financing.

In creating a bias in favor of producers, developing countries have stunted the development of their financial systems. This makes it hard for them to use foreign financing to expand domestic demand for goods and services effectively. Indeed, with the exception of foreign direct investment—for instance, Toyota’s setting up its own factory in China—foreign financing ultimately relies, either directly or indirectly, on the willingness of the developing-country government to support domestic expansion, rather than on the ability of its private sector to do so. Because foreign lenders focus on the creditworthiness of the country and its government rather than on the specific attributes of the project being financed, and because they effectively obtain seniority over domestic lenders, foreign lending tends to be more permissive than it ought to be, given the benefits of the project. Furthermore, because political compulsions invariably force governments to press hard on the accelerator, countries tend to overuse foreign finance until they are yanked back by a sudden stop in foreign inflows. The ensuing bust tends to set back growth tremendously.

A government-directed, producer-biased strategy of growth tends to stunt the development of that country’s financial sector. Because banks are told whom to lend to, and because domestic competition among producers is limited anyway, banks tend not to seek out information or develop their credit-evaluation skills. The legal infrastructure to close down weak borrowers, or to enforce repayment from recalcitrant ones, is virtually nonexistent.

As we saw earlier, the government does try to help producers by setting deposit rates low in order to lower the cost of credit. However, because interbank competition is limited, banks tend to become very inefficient, with bloated staffs and excessively bureaucratic procedures. Anyone who wishes to cash a check at a public-sector bank in a developing country would do well to develop an attitude of resignation while watching the check crawl from desk to desk, adding signature upon signature, before it finally appears at the cashier’s window. These inefficiencies as well as limited competition result in an enormous interest spread (the difference between the bank’s lending rate and its cost of funds): in Brazil, the spread routinely has been more than 10 percentage points even for short-term loans to corporations in good standing, whereas it is a fraction of a percent in industrial countries. Thus much of the cost of capital advantage obtained by setting deposit rates for households very low is lost through inefficiencies in the banking sector.

Not only do households get little for their savings, but they also find it very difficult to borrow. Lending to households is very risky even in a modern financial system like that of the United States, and doubly so in an underdeveloped financial system. Mechanisms to track credit histories simply do not exist. Because few people have jobs in the formal documented sector, and a significant portion of incomes—such as remittances sent by workers to their parents—are not based on formal contracts, banks have little information on which to base lending decisions. And because the judicial system does not allow easy enforcement of claims against assets, banks cannot lend easily against houses or washing machines.

Households do borrow from moneylenders, with kneecaps sometimes serving as collateral against default, but such borrowing takes place at astronomical rates of interest. The formal banking system could charge high interest rates to compensate for the risk of default (but lower rates than the informal system), but such practices are typically blocked by politicians, who “fight” for citizens by capping formal-sector interest rates at low levels, thus making lending unprofitable and driving households to the informal and unregulated moneylenders. Everything changes as the financial system develops, but this is a reasonable description of many developing countries’ financial systems in the 1990s.

I said earlier that there were broadly two reasons a developing country could need foreign financing: if it saved little relative to the norm or invested significantly more than the norm. Let us now consider circumstances in which a country saved too little.

Given that developing-country households find it hard to obtain retail credit, they typically cannot overrun their budgets. Instead, they save for a rainy day, anticipating the difficulties they will experience when times get tough. Unlike its counterpart in the United States, therefore, the underdeveloped financial sector in a developing country makes it difficult for the government to use easy credit as an instrument to carry out populist policies. Instead, the developing country’s government performs the role of the financial sector, borrowing and offering transfers and subsidies to favored constituencies so as to allow households to spend more. So the primary reason a developing country saves little is that the government runs large deficits and borrows to finance them. Usually, these deficits are caused by overspending on transfers and subsidies to politically favored segments of the population—by political logic rather than economic rationale.

For example, Kenya, a country that survives on international aid and had an annual per capita income of US$463 in 2006, paid its legislators a base compensation of about $81,000 a year, tax free, plus a variety of allowances and perks that effectively doubled their take-home pay.
⁵
In a year of widespread drought, a favored car for legislators was the Mercedes-Benz E class, supported by a “basic” monthly car allowance of $4,719. The legislators had the gall to hold up a drought-relief bill as they demanded a higher car allowance, citing the shoddy condition of the roads in their constituencies as justification. Unsurprisingly, the public’s demand that the legislators fix the roads had little effect, but the raise went through. These “public servants” earned significantly more than most Kenyan corporate executives and more than their counterparts in the developed world.

When the government’s borrowing exceeds available domestic resources, and it does not have access to official government-to-government aid, it turns to foreign private lenders. Because governments command some credibility, and have some access to borrowing from multilateral institutions like the IMF, lenders are willing to finance them for a while. Knowing, however, the temptations that face an opportunistic government—to inflate away debt if denominated in the domestic currency or to pile on more and more debt on existing debt, thus eroding its value—foreign lenders take precautions. They demand repayment in foreign currency (which is not affected by the country’s ability to inflate or devalue its currency), and they shorten the terms of their loans in proportion to the country’s indebtedness, so that they can pull their loans at short notice.

Instead of controlling its spending, therefore, the populist government that has exhausted its ability to borrow domestically turns to foreign lenders to finance it. Thus the circumstances in which foreign loans are made are not propitious. Knowing this, foreign lenders demand protection, which the government can typically give only by eroding the rights of existing domestic creditors—for instance, the more the overindebted government borrows in foreign currency, the higher the inflation it will eventually have to generate to erode domestic-debt claims on it. Moreover, because foreign investors make short-term loans, any adverse political development may scare them into refusing to refinance the government. Even more problematic, a substantial improvement in opportunities in their home countries—such as a rise in interest rates—can cause foreign investors to pull their money out en masse.
⁶

All these factors were at play in the Mexican crisis of 1994. The
sexenio,
or six-year administration, of President Carlos Salinas de Gortari was coming to an end. In the traditional fashion of the dominant party, the Partido Revolucionario Institucional (PRI), he launched a spending splurge to keep voters happy. Domestic savings dropped by 3.3 percent of GDP between 1991 and 1994, with much of it accounted for by an increase in government spending, leading to a current-account deficit that touched 7 percent of GDP in 2004. Even while the need for foreign financing mounted, political developments took a turn for the worse. In Chiapas, aggrieved peasants rose up in an armed rebellion against the government, and later in the year, the PRI’s presidential candidate, Donaldo Colosio, was assassinated. Moreover, the Federal Reserve of the United States raised interest rates throughout 1994, from 3 percent to 5.5 percent, giving investors the incentive to bring their money back to the United States.

As foreign investors became more worried about financing the Mexican current-account deficit, the government started converting its short-term peso-denominated debt into
tesobonos
—short-term bonds that were indexed to the dollar-peso exchange rate and would protect investors from a devaluation. But as political uncertainty increased, even this was not protection enough. Investors started selling out, converting their pesos into dollars, and departing the country. The central bank’s exchange reserves became depleted, and the new president, Ernesto Zedillo, had a full-blown crisis on his hands when he entered office. Eventually, an enormous loan was put together by the U.S. Treasury and the IMF to prevent Mexico from defaulting on its debts. Investors in the
tesobonos
were paid back, but the country went through a wrenching crisis, and those who held on to peso-denominated debt suffered heavy losses.