| Econ. 153a | C. Sims |
You are to pick one of the topics below or get an alternate choice of topic approved. Your paper should be about 5 pages of double-spaced text, plus plots that support discussion in your text. Papers more than 10 pages long are not welcome. If you really do so much work you need that amount of space, you can submit a longer paper, but there is no extra credit for extra length -- indeed papers longer than is justified by their substance lose credit for that. There is no minimum length to the paper. If you think you can write a 3-page paper that looks good compared to others' 5-page papers, go ahead.
For most of the topics below at least some relevant data appear in Citibase, which is accessible in Statlab, or on the net in FRED, which you already know how to access. More extensive foreign data appears in the IFS database, also available at Statlab. IFS stands for International Financial Statistics, an International Monetary Fund publication that provides the same data in print form. The printed IFS publication is available in the Social Science Library and serves as a kind of catalog for the IFS databank. Another print source for foreign data, not available on a computer databank at Yale, is the monthly OECD (Organization for Economic Cooperation and Development) publication Main Economic Indicators. They occasionally put out historical summary volumes as well. You are not limited to these sources, of course; indeed you are encouraged to come up with relevant data from sources of your own. Though it is certainly handier to use a database, entering a few hundred numbers by hand from a printed table only takes a few minutes.
The questions listed under some of the topics may be more than you can handle well in 5 pages. Feel free to focus your paper on a subset of the questions posed if that will make it better.
Canada, Mexico and the U.S. are tightly linked as trading partners, and were so linked even before NAFTA. The neoclassical growth theory that predicts convergence of economies to a common growth path over time might seem to apply to them. What are the facts? Are the income levels becoming closer, at least proportionately? It can be argued that because of fears of "political risk" (the risk of government action to expropriate all or part of foreign-owned assets) U.S. investment in Mexico has been more inhibited than U.S. investment in Canada. Is there evidence for this? Free cross-border investment should speed convergence. (Your paper would explain why.) Is there evidence for this in the comparison of Canadian and Mexican experience?
How much did U.S. government debt increase, in real terms, during the Reagan administration? Is it still increasing? How does this increase compare with previous increases in U.S. history? How does the increased debt compare with the total wealth of the private sector? What is the range of uncertainty about the correct measurement of the deficit and of public debt? Is there evidence that private savings in other forms were crowded out? Did real interest rates show effects of the changes in debt?
Inflation reduces the value of dollar-denominated public debt. Effects on the wealth of holders of short-term interest-bearing debt are small, because interest rates quickly adjust to offset any sustained inflation. Holders of long-term debt, though, can suffer repeated losses from inflation over several years, as can holders of cash. Find out how much U.S. government debt was in various maturity classes at the beginning of the inflationary period of the 70's, and calculate the losses incurred by the holders of that debt due to inflation, up to the time of maturity. If you can, try to determine how much of this loss was borne by foreign holders of U.S. debt. Who gained at the expense of those who suffered these losses?
It has been proposed that the US switch from "pay-as-you-go" social security to a fully funded scheme, in which the social security tax, instead of going into the government's general revenue pool, would be used to purchase private capital. Workers would then be given the value of the capital that had accumulated upon retirement, as an annuity. Find out how large social security is as a part of private saving. Use an OG model with two-period lived agents to assess quantitatively the likely impact on the welfare of the first generation whose retirement is fully funded. (This will vary according to how taxes to finance the reform are raised.)
At least by some measures, real wage growth in the U.S. slowed down dramatically or even ceased beginning sometime in the 1970's. The story is somewhat different for family income. A possibly related fact is that wage inequality between high- and low-paid workers and across skill classes has been increasing. Find data to document these facts for the U.S. You might try to obtain data for other countries to see the degree to which they show similar patterns. Discuss possible explanations, and how well they accord with the data. To the extent possible, use a growth model with labor input as a framework for your discussion.
References: The papers in the Feb. 1992 issue of the Quarterly Journal of Economics, especially those by Katz and Murphy and Murphy and Welch, the paper by Cutler and Katz in the 1992:2 issue of Brookings Papers on Economic Activity, and a paper by Frank Levy that may have appeared by now in the Journal of Economic Literature and if not is available as a Harvard discussion paper (for which check the Cowles library at 30 Hillhouse.)
Monetarism is the theory that monetary policy can be measured by observed movements in the stock of money and that erratic monetary policy generates much or most of observed business fluctuations. In an ISLM model it can be characterized as the theory that LM is steep, IS is not steep, and LM shifts around a lot because of erratic monetary policy while IS stays relatively stable. Much of this theory's appeal is based on the observed co-movements of money stock and income or GNP in historical time series. There is a vast literature on this subject, with many statistical studies.
It has been somewhat less active recently, however, and you might check whether regularities claimed in the older literature have persisted into recent years. How sensitive are conclusions to the choice of a measure of "money" (e.g. M1, M2 , M3 ) and how are these measures defined? Do the data behave the same in other countries as in the U.S.? In particular, you might look at how well the patterns asserted in the Friedman and Schwartz Monetary Trends... book held up after Margaret Thatcher's accession to power in the U.K., as her government tried to implement monetarist ideas.
References: Milton Friedman and Anna Schwartz, A Monetary History of the United States 1867-1960 (a long book -- you might read a part of it related to a particular historical period, e.g. the Great Depression or 1947-60); James Tobin, "Money and Income: Post Hoc ergo Propter Hoc?", Quarterly Journal of Economics 1970; Friedman and Schwartz, "Money and Business Cycles," Review of Economics and Statistics 1963 (supplemental issue); Monetary Trends in the U.S. and the U.K.: Their Relation to Income, Prices and Interest Rates 1867-1976 (another long book); Finn Kydland and Edward Prescott, "Business Cycles: Real Facts and a Monetary Myth" in the Spring 1990 Quarterly Review of the Federal Reserve Bank of Minneapolis.
Years ago Phillips found a relation between wage inflation and unemployment in the U.K. In the 60's, such a relation seemed to exist in the U.S., but it apparently disappeared in the 70's. Has it returned recently? Can you find it in data for other countries? Is unemployment more strongly related to the real wage than to the nominal wage? Modern theories of inflation emphasize the importance of expectations. You should discuss how such theories relate to interpretation of the data you display.
Most of macroeconomic data is used in "seasonally adjusted" form, meaning that regular patterns of variation across seasons of the year are removed before it is published. But "seasonal cycles" may not be completely distinct from "business cycles". Are seasonal patterns in macroeconomic variables similar to cyclical patterns? E.g., was there a "seasonal Phillips curve" in the U.S. during the period in the 50's and 60's when there seemed to be a Phillips curve in annual data? Are money and income related in their seasonal patterns the way they are at longer time intervals? Refer to "The Economics of Seasonal Cycles" by Jeffrey A. Miron (a discussion paper), which considers these issues, and/or to several other papers in the last few years by Miron and co-authors.
Systematic errors in measuring price change usually lead to systematic errors in measuring output change, and thus productivity change. Two leading examples of potentially severe problems in measuring price change are for computers and for services (like retail trade, medical services, financial services). Discuss how price changes are actually measured for computers and one or more service industries, suggesting how large the errors might be and what the implications might be for measurement of productivity growth in the industry to which the price applies and in other industries. Refer to "The Productivity Slowdown, Measurement Issues, and the Explosion of Computer Power," by Martin Neil Baily and Robert J. Gordon in the Brookings Papers on Economic Activity, 1988:2.
The Japanese economy has grown very fast since the end of World War II and has shown a much higher savings rate than the U.S. economy. Are these commonly understood facts really firmly established? If so, are they explainable as a Japanese reaction to wartime devastation, with a likely convergence of Japanese to U.S. growth patterns as time goes on? Use the models containing investment and capital that we discuss in this course. The Japanese age distribution is shifting upward rapidly, and they rely less on pay-as-you-go government social security than does the US. How much of the difference between Japanese and US savings is accounted for by these facts? [The age distribution question, which is not discussed much by Christiano, may be a paper topic of its own.] See the article by L. Christiano in the Spring 1989 Quarterly Review of the Federal Reserve Bank of Minneapolis and the article by F. Hayashi to which Christiano refers.
At this time the Federal Reserve Board announced a drastic shift in policy, apparently toward greater reliance on monetarist principles. The shift is clearly visible in at least some of the monthly and weekly monetary time series. What happened to the behavior of monetary aggregates and interest rates? Is it what should have been expected from the nature of the announced policy change? Did people at the time expect the changes in volatility of interest rates and of monetary aggregates that occurred? Should we have learned anything about monetarism from the episode? An interesting angle might be to compare this episode with what happened in the U.K. in the period after Thatcher instituted a somewhat different version of monetarist policy.
The Federal Reserve System was established in 1914. Monetary aggregates and interest rates thereupon started behaving differently. What was the Fed trying to do? Did it succeed? Did it make things better, and in what sense?
References: Jeffrey Miron, "Financial Panics, the Seasonality of the Nominal Interest Rate, and the Founding of the Fed," American Economic Review 1986; Miron and N. Gregory Mankiw, "Should the Fed Smooth Interest Rates? The Case of Seasonal Monetary Policy," Carnegie-Rochester Policy Series number 34, 1991. These references focus on seasonality, but themselves contain references to broader discussions of the founding of the Fed.