Summaries of Recent Research

Most economics research at MIT focuses on tangible economic problems from both a theoretical and an empirical perspective. Empirical research may test theoretical models of economic behavior, or it may seek to evaluate whether current policies are effective, ineffective, or counterproductive in achieving their goals. Theoretical research may either establish new conceptual frameworks for studying markets and economic institutions, or lead to new statistical and analytical tools.

Business and government decision makers in the U.S. and abroad frequently seek out MIT faculty for help in formulating and evaluating business decisions and economic policy initiatives. The faculty looks forward to continuing leadership in business and public policy.

Understanding Global Imbalances
by Ricardo Caballero, Ph.D.
Ford International Professor of Economics, MIT

Research Highlight No. 1

The global economy has been characterized in recent years by striking imbalances in capital flows, a continuous flow of capital from poorer countries to richer ones, volatile asset markets, several “speculative bubbles” that have driven asset prices up and then burst, and historically low real interest rates. A variety of explanations have been advanced for these disparate phenomena. But a new paper by MIT Economist Ricardo Caballero, building on his joint work with several other MIT economists, has a simple explanation for all of them: a shortage of global assets. The world economy is saving as always, but there are not enough assets in which to invest this savings. Read more...

Explaining the Persistence of Disagreement in Bargaining
by Daron Acemoglu, Ph.D., Charles P. Kindleberger Professor of Applied Economics
Victor Chernozhukov, Ph.D., Associate Professor of Economics
Muhamet Yildiz, Ph.D., Pentti J.K. Kouri Career Development Associate, Professor of Economics

Research Highlight No. 2

Economic transactions are frequently characterized by disagreements between the participants. The buyer and seller of a common stock typically disagree either about the future prospects for the company it represents or about the returns available from alternative investments. The various bidders in a spectrum auction, or in an auction for timber or oil and gas rights, disagree about the potential profit opportunities from exploiting these resources. Explaining disagreement is therefore an essential component in most economic models of decision-making and market interaction. Nearly all existing models assume that individuals have consistent beliefs about their economic environment and about the nature of the payoffs from each decision they might make. In any actual decision, different actors may have different information about the specific payoffs in question. This may lead them to different decisions. Hypothetically, however, if they all had access to the same information, and if the information set included a large amount of data on the relevant payoffs, then their evaluation of the alternative choices would converge. This is the meaning of consistency in decision-making models. Consistency is often justified by noting that common experiences and observations will eliminate major disagreements about the economic environment and the nature of various payoffs, even if specific disagreements about payoffs might occur in the presence of limited information. Read more...

EZ-Tax: The link between the visibility of taxes and the level of taxation
by Amy Finkelstein, Ph.D.
Associate Professor of Economics

Research Highlight No. 3

How much did you pay in tolls the last time you drove on a toll road? You probably paid using electronic toll collection such as EZ-Pass in much of the Northeastern United States, and Fast-Trak in the Bay Area. As a result, you may have very little sense of current toll rates. Read more...

Oil Shocks: They Ain't What They Used to Be
by Olivier J. Blanchard, Ph.D.
Class of 1941 Professor of Economics

Research Highlight No. 4

There have been two periods of large and sustained increases in oil prices during the last four decades. The first took place in the 1970s, and the real increase in oil prices came with large increases in inflation and a sharp economic slowdown. The second started in the late 1990s and has continued through the current decade. Yet, during that time, growth has been sustained, and inflation has remained low. Why? In a new paper, “The Macroeconomic Effects of Oil Prices: Why are the 2000s so Different from the 1970s?,’’ Class of 1941 Professor Olivier Blanchard (Ph.D. ’77) and Jordi Galí (Ph.D. ’89), Director of CREI in Barcelona, offer a series of explanations. Read more...

 

Building Dams: Who Wins and Who Loses

The construction of dams is one of the most costly and controversial forms of public infrastructure investment in developing countries, yet surprisingly little is known about the economic impact of such projects. In “Dams,” which has just been published in the Quarterly Journal of Economics, Esther Duflo and Rohini Pande of Harvard University examine the evidence on the effects of dam building on surrounding areas. They find that the building of large dams in India has had significant distributional impacts across areas. The areas where the dams were built saw a rise in poverty, while areas downstream from dams saw agricultural productivity rise and poverty fall. In aggregate, dams appear to have slightly raised productivity, but this effect is small relative to the significant redistribution across areas induced by their construction.

Shill Bidding and Optional Auctions

The analysis of auctions has been one of the most active topics of research in economic theory during the last three decades. This body of work has delivered important insights on the similarities and differences between different types of auctions, and has shown that auctions have remarkable revenue and efficiency properties. Because auctions may vary on so many dimensions, however, there are still new issues to explore. In “Shill Bidding and Optimal Auctions,” Sergei Izmalkov investigates the effect of allowing the seller to participate as a bidder in an ascending-price auction, such as the one being used by eBay. When the seller can submit bids that buyers must consider and respond to, and when the item does not sell unless the potential buyers’ bids exceed the bids submitted by the seller, Izmalkov shows that the standard ascending-bid auction maximizes the revenue of the seller among all possible ways to sell. This is not achievable in general by other auction forms, even with reserve prices. This finding may explain why sellers in on-line auctions are willing to engage actively in bidding.

Improving Estimation Methods

Instrumental variable (IV) estimators, which offer a powerful tool for identifying causal relationships in markets and other settings with multiple causal pathways, are widely and increasingly used in applied econometric research. In some settings, however, IV estimators may break down—for example when there is heteroskedasticity and the instrument set is large. In a new paper on econometric methodology, “IV Estimation with Heteroskedasticity and Many Instruments,” Jerry Hausman, Whitney Newey, and Tiemen Woutersen of Johns Hopkins University provide new insights on the best way to address some of the limitations of IV estimation. They compare several different estimators that might be used in this setting, rank their performance, and offer applied researchers a set of new tools that will lead to reliable statistical inference even in this challenging setting.

Will Global Warming Harm Agricultural Productivity?

Policy makers in many nations are increasingly interested in calibrating the potential economic consequences of global warming. The agricultural sector plays a central role in such analyses, since agricultural output depends so heavily on the weather. Yet the impact of global warming on agricultural productivity is highly location-specific and often ambiguous. While rising temperatures affect farm productivity, so does variation in rainfall—and warming may induce rainfall changes. Michael Greenstone and UC-Santa Barbara researcher Olivier Deschenes provide new evidence on how global warming may affect agricultural productivity in “The Economic Impacts of Climate Change: Evidence from Agricultural Output and Random Fluctuations in Weather.” They note that the U.S. agricultural sector has considerable history with the type of weather fluctuations projected to arise from global warming. By studying how past weather fluctuations affected agricultural output, they conclude that the changes projected to arise from global warming are likely to have a surprisingly small impact on agricultural output. Their estimates suggest significant variation across areas, however, with a large negative effect on agricultural profits in states such as California and North Carolina,and a large positive effect in states such as North Dakota.

Zombie Lending in Japan

A working paper by Ricardo Caballero, Takeo Hoshi (Ph.D. 1988) of the University of California, San Diego, and Anil Kashyap (Ph.D. 1989) of the University of Chicago Graduate School of Business analyzes the role of the Japanese banking sector in contributing to Japan’s stagnant economy during the 1990s. In “Zombie Lending and Depressed Restructuring in Japan,” they observe that lax banking regulation allowed Japanese banks to comply with capital standards throughout the decade even though they would have failed to meet these standards if their loan portfolios were marked to market. To forestall bankruptcies at borrowing firms that were no longer able to repay earlier loans, banks continued to extend credit, thereby supporting these “zombie firms” and allowing them to remain in business. The presence of these firms reduced the profitability of the firms that were solvent and had promising futures, reduced the incentives for the successful firms to invest, and prolonged Japan’s slow recovery after the asset market melt-down of the early 1990s.