Search results

Economists have developed models in which individuals form expectations of key variables in a 'rational' manner such that these expectations are consistent with actual economic environments. Professor Sheffrin first explores the logical foundation of the concept and the case for employing it in economic analysis. Subsequent chapters investigate its use in macroeconomics, financial markets, and microeconomics. A final chapter assesses its impact on theoretical and empirical work in economics and policy arenas. The author argues that while rational expectations are still central to macroeconomic policy debates, fully workable models have not yet been devised, and offers reasons for the lack of practical and conceptual progress. All chapters of the second edition have been revised or expanded. New sections inter alia include material on learning, the rationality of reported expectations, alternative recent developments explicitly or implicitly using rational expectations, new tests of the Lucas critique, and models of noise trading. The book is written in a non-technical fashion for beginning graduate students and non-specialists

Followers of law, politics and business commonly relate stories of individuals who appear to predict an expected performance level below what they believe themselves to be capable of. The standard explanation for such rhetoric is that it hedges against the negative consequences of unanticipated failures and takes advantage of unexpected successes. Although the strategy appears highly attractive, some individuals do provide honest evaluations of their abilities, and some overpromise. We develop a model of strategic communication designed to explain this variation. Underpromising is especially attractive when observers have strong incentives to watch a preliminary performance; however, when high-quality individuals are in large supply and when the costs of performing badly are neither too high nor too low, underpromising can result in individuals being ignored. To ensure that they are not, individuals must give up the opportunity to outperform a promise and risk an underperformance. [Reprinted by permission of Sage Publications Ltd., copyright holder.]

We introduce a St. Petersburg-like game, which we call the 'Pasadena game', in which we toss a coin until it lands heads for the first time. Your pay-offs grow without bound, and alternate in sign (rewards alternate with penalties). The expectation of the game is a conditionally convergent series. As such, its terms can be rearranged to yield any sum whatsoever, including positive infinity and negative infinity. Thus, we can apparently make the game seem as desirable or undesirable as we want, simply by reordering the pay-off table, yet the game remains unchanged throughout. Formally speaking, the expectation does not exist; but we contend that this presents a serious problem for decision theory, since it goes silent when we want it to speak. We argue that the Pasadena game is more paradoxical than the St. Petersburg game in several respects. We give a brief review of the relevant mathematics of infinite series. We then consider and rebut a number of replies to our paradox: that there is a privileged ordering to the expectation series; that decision theory should be restricted to finite state spaces; and that it should be restricted to bounded utility functions. We conclude that the paradox remains live.

We introduce a St. Petersburg-like game, which we call the 'Pasadena game', in which we toss a coin until it lands heads for the first time. Your pay-offs grow without bound, and alternate in sign (rewards alternate with penalties). The expectation of the game is a conditionally convergent series. As such, its terms can be rearranged to yield any sum whatsoever, including positive infinity and negative infinity. Thus, we can apparently make the game seem as desirable or undesirable as we want, simply by reordering the pay-off table, yet the game remains unchanged throughout. Formally speaking, the expectation does not exist; but we contend that this presents a serious problem for decision theory, since it goes silent when we want it to speak. We argue that the Pasadena game is more paradoxical than the St. Petersburg game in several respects. We give a brief review of the relevant mathematics of infinite series. We then consider and rebut a number of replies to our paradox: that there is a privileged ordering to the expectation series; that decision theory should be restricted to finite state spaces; and that it should be restricted to bounded utility functions. We conclude that the paradox remains live.

 A lovely plot from the always interesting Torsten Slok. The graph shows the actual federal funds rate, together with the path of "expected" funds rate implicit in fed funds futures market prices. (Roughly speaking the futures contract is a bet on where the Fed funds rate will be at various dates in the future. If you want to bloviate about what the Fed will do, it's easy to put your money where your mouth is!) A lot of graphs look like this, including the Fed's "dot plot" projections of where interest rates will go, inflation forecasts, and longer term interest rate forecasts based on the yield curve (yields on 10 year bonds imply a forecast of one year bonds over the 10 year period.) Just change the labels. In words, throughout the 2010 zero bound era, markets "expected" interest rates to lift off soon, year after year. It was sort of like spring in Chicago -- this week, 35 degrees and raining. Next week will be sunny and 70! Rinse and repeat. Once rates started rising in 2016, markets actually thought the rise would be slower than it was, but then did not see the end of the rise. Of course they did not see the sudden drop in 2020, because they didn't see covid.  I find it fascinating that for the first full year of inflation, 2021-20222, markets did not price in any interest rate rise at all. The Taylor rule (raise interest rates promptly when inflation rises) wasn't that forgotten at the Fed! The one time when it made abundant sense to forecast the Fed would raise rates, markets did not reflect that forecast. When the Fed finally did start to raise rates, amid raging inflation, the market even more curiously thought the rate rises would stop quickly. This being a pasted graph, I can't easily add inflation to it, but with the federal funds rate substantially below inflation until June 2022, it's interesting the markets thought the Fed would stop. The story of "transitory" inflation that would go away on its own without a repeat of the early 1980s -- without interest rates substantially below inflation -- was strong. The market forecast seems to me still remarkably dovish. GDP just grew like gangbusters last quarter, and the Fed believes in the Phillips curve (strong growth causes inflation). We're running a historic budget deficit for an economy at full steam. The Taylor rule (interest rates react to inflation and output) is still a pretty good description of what the Fed does, sooner or later.  So, if you were to trade on the historical pattern, you would bet on rates falling much more quickly than forecast. Hmm. This is an old phenomenon. The "expectations" in market forecasts don't seem right. Don't jump to fast to "irrational," finance always has a way out. We call it the "risk premium." There is money to be made here, but not without risk. If you always bet that the funds rate will be below the futures rate, you'll make money most of the time, but you will lose money on occasion. First, in many such bets the occasional losses are larger than the small regular gains. That is important, because the pattern of constant misses in the same direction suggests irrational forecasts, but that's not true. If you play roulette and bet on anything but 00, you win most of the time, but lose big on occasion and come out even overall, More  plausibly, when you lose you lose at times when it is particularly inconvenient to lose money. Economists often use the federal funds future to establish the "expected" federal funds rate, and then any movement including no movement at all counts as an "unexpected" shock. By that measure the early 2010s were one series of "unexpected" negative monetary policy shocks, month after month. The graph makes it clear that's a reading of history that needs some nuance in its interpretation.