Capital controls—for better or worse?
Can limits on international capital flows be beneficial? Yes, if conditions are right
Published March 31, 2016
Capital controls—taxes and other limits on international transactions in financial assets—have been used by governments since World War I, when major warring nations restricted the purchase of foreign assets or loans. Controls helped finance war expenses by keeping capital at home where it could be taxed; by reducing interest rates, controls also enabled governments to borrow more cheaply. Their use has waned and waxed since the end of the Great War, surging during the Great Depression and World War II, and was explicitly endorsed at the 1944 Bretton Woods conference as a means of easing balance of payments problems by stemming capital outflows.
But their costs were also evident. They distort the allocation of investment, preventing the international flow of capital to where it’s most productive. They prevent nations from softening natural disasters and economic downturns with international loans. Just as tariffs and embargoes limit welfare gains from international trade in goods during a single time period, so capital controls impede gains from trading goods in the present in exchange for goods in the future. Since Margaret Thatcher’s Britain lifted its exchange controls in 1979, the pendulum has swung strongly in favor of free international capital flows.
Are controls beneficial?
The pendulum is swinging back. In the past five to 10 years, more nations have begun to consider and/or adopt capital controls, worried by—among other things—the instability introduced by foreign financial firms and institutions that invest and withdraw at will. Iceland, for instance, was hard hit when foreign investors yanked their funds during the 2008-09 financial crisis. By contrast, Switzerland has been hurt by a surge in foreign inflows that strengthen their franc, contribute to deflation and hurt Swiss exporters. Even the International Monetary Fund, long a staunch advocate of free capital flows, has endorsed the limited use of controls.
“On the Desirability of Capital Controls,” a Minneapolis Fed Staff Report (SR 523) by monetary advisers Jonathan Heathcote and Fabrizio Perri, examines the question, focusing on the impact of controls on interest rates and the terms of trade. It concludes that curbing the free movement of capital can, in theory, change prices of both money and products in favor of a nation that adopts controls. In some environments, bilateral capital controls can be beneficial.
The mechanism behind this finding calls for a bit of explanation. First, consider a hypothetical country without capital controls. If it enjoys a positive productivity shock—to be concrete, suppose it invents a very efficient electric car—it will first borrow internationally to invest in new facilities to produce the car. Over time, the country will use part of the proceeds from the sales of the new car to repay its international loans; some of the rest will be saved abroad to finance future expenditure needs.
Now, if that same country had instead crafted capital controls that taxed foreign borrowing during the first phase (when the nation is borrowing from abroad to invest in its newfound productivity) and that taxed savings during the later stage (when it saves abroad), it would tend to borrow less in the short term and save less in the long term. The controls would clearly have a negative impact: Productivity-enhancing investment will be underfunded, and consumption volatility will be higher than otherwise. “By itself, this would be detrimental to consumption smoothing and the efficient allocation of capital,” write Heathcote and Perri.
Interest rates and terms of trade
The previous analysis ignored one important variable: price. In particular, how do capital controls affect interest rates (the price of borrowing and lending) and the terms of trade (the price at which domestic firms can sell this fancy new car abroad)?
The impact on interest rates? Capital controls that reduce borrowing from abroad initially will drive down the equilibrium world interest rate in this phase. And by reducing foreign lending in the later stage, capital controls will drive up the world interest rate when the country is a foreign lender. Thus, by imposing capital controls, a country can borrow more cheaply when it is borrowing and lend more dearly when it is lending. “These changes in interest rates potentially leave the country better off,” the economists write.
As for the terms of trade, a country enjoying a productivity surge will, again, borrow internationally to increase its investment. More borrowing allows greater investment and leads to increased supply of the new car. But greater supply lowers the price of the new car, potentially damaging the nation’s terms of trade. Capital controls limit this damage. Because they curb the initial borrowing that a country might undertake to benefit from its productivity shock, controls “dampen this terms of trade response,” write Heathcote and Perri, “a second possible motivation” to adopt them.
The economists build a standard business cycle model with two hypothetical nations that allows them to analyze and quantify how this theoretical mechanism of capital controls affects the impact of productivity shocks on quantities, prices and economic welfare. They then put it to work in several scenarios: (1) a single nation adopts capital controls, (2) two trading partners adopt controls and (3) a benevolent planner designs a scheme to maximize benefits for both nations.
When the economists analyze a scenario in which a single country adopts capital controls (in the form of a tax on capital inflows and outflows), they find that controls provide definite benefits. When one country adopts the tax and its trading partner doesn’t, the economists calculate, that “country will find it desirable to impose a substantial tax” on capital flows—one that will cut flows of capital across borders by a factor of three relative to no controls.
How about a country so small it couldn’t possibly alter international interest rates? It too will want to adopt strong controls, as long as it produces a good whose terms of trade it can affect.
“Capital control wars”
A nation faced with capital controls imposed by its trading partner is likely to respond in kind. If both nations set their capital taxes optimally to maximize their respective citizens’ welfare, taking as given the other nation’s policy—a scenario they call “capital control wars”—the economists find that both nations will set the capital tax at 16.1 percent, and intertemporal trade will plummet. The average net foreign asset position will fall from 45 percent of GDP to under 3 percent, and “both countries suffer a welfare loss.” Declaring a ceasefire and allowing for the free flow of international capital would benefit both nations.
Can we all benefit?
Given that no nation would stand idly by as its trading partner adopted a capital control, does the “war” scenario suggest that controls are never beneficial? The answer, the economists find, is no, at least under certain conditions. “Symmetric capital controls imposed in both countries can be welfare improving,” they write.
Why could properly designed controls enhance welfare for both nations, and under what conditions? The intuition, and perhaps the most novel contribution of the paper, is that capital controls can improve international risk sharing when international financial markets don’t provide adequate insurance. When a nation experiences a negative productivity shock, for instance, its terms of trade will improve, thereby providing a measure of automatic insurance. How much the terms of trade will move depends on how easily countries can borrow and lend. Thus, by adjusting the level of capital controls, terms of trade insurance can potentially be improved.