Economist's Commentary: April 24, 2008

Off-the-Mark Forecasts

By Jed Smith, Managing Director, Quantitative Research

Forecasting attempts to predict the future — i.e., the economic outlook, future business conditions, product opportunities, and possibly a myriad of social, economic, political, and other conditions. By itself a forecast is a possibly interesting piece of trivia unless someone is going to take some type of action based on the forecast. That is, the relevant question to ask concerning the forecasting process is "What would we do differently if we had some insight about the future?" 

Therefore, forecasts are useful if they improve the decision making process. A decision to ignore forecasting on the assumption that "all forecasts are bad" is also a decision to accept random fate. A bad forecast is better than no forecast if there is some improvement from whatever choices we would have made if we didn't have a forecast. This article will provide some perspective on the forecasting process and some suggestions on coping with forecasting accuracy.

The first economic forecast, recorded in the Book of Exodus as seven fat years followed by seven lean years, was on-the-mark. The information was then used for agricultural planning purposes. Forecasting experience has subsequently deteriorated:

• Economist Irving Fisher — still noted today based on his pioneering work in monetary theory — was widely quoted in 1929 that stocks had reached what looked like a permanently high plateau. The market crashed along with his money and reputation.
  
• Thomas Watson, President of IBM, initially forecast a total market potential of five computers. Experts make mistakes.
  
• The initial target for the Intel 8088 microprocessor that powered IBM's first PC projected target sales of 100,000 per year. Marketing people make mistakes.
  
• John Unitas, Joe Montana, and Tom Brady, were all relatively low draft choices: Coaches make forecasting mistakes.
  
• Approximately 20 publishers turned down the first Harry Potter volume by J.K. Rowling. The Dr. Seuss book And to Think That I Saw It on Mulberry Street was rejected 27 times. Publishers make mistakes.
  
• Monkey dart throws often outperform stock picks by hedge fund managers.
  
• And the Best One: Decca Recording rejected the Beatles — didn't like the sound, and guitar music was on the way out.

Economic forecasts are frequently a bit off-the-mark. Scott Shuh, an economist with the Boston Federal Reserve Bank, has documented that over the 1995 –2001 time frame the average GDP forecast missed by up to 2 percent, with unemployment rates typically forecasted by half a percent higher than actually occurred. Congressional Budget Office and other government forecasts along with other studies published in conjunction with the Wall Street Journal or the National Association for Business Economics often illustrate the inaccuracy of economic projections.

Why are Economists frequently wrong? Forecasts are based on assumptions and modeling about government, Federal Reserve, consumer, and business behavior. However, any forecasted outcome can be significantly changed by unexpected and unknowable exogenous shocks — e.g., changes in government or monetary policies, changes in consumer confidence, unexpected business investment, changes in international trade, and a wide variety of other unforeseen possibilities.

A variety of useful forecasts are available — from the Philadelphia Federal Reserve, from Wall Street Journal compilations, and a variety of commercial sources. How does one measure and evaluate the success of forecasts? That is, what is the appropriate success rate for a forecaster? In baseball anything above three hits in ten at-bats is regarded as successful. In 3-point attempts in basketball anything above four goals in ten shots is excellent. A good quarterback completes more than five in ten forward passes. Although the NAR forecast has been off-the-mark at times, it has been less wrong than other forecasts. USA Today has ranked NAR quarterly forecasts in the top 10, based on an analysis by the Federal Reserve Bank of Atlanta.

The value of forecasting can be seen in light of current economic conditions as of April 2008. Forecasts bring some insight — although possibly imperfect — to the analysis of economic conditions. Economists currently vary widely on the subject of whether the economy is or will be in a recession, whether unemployment will be 5.2 percent or 5.5 percent (or whatever), whether interest rates will decline or increase, and whether inflation will increase by half a percent or one percent. This may look like a lot of uncertainty. It isn't. Economists agree that the economy is slowing down for awhile but will pick up in 3 quarters, more or less. Therefore, a business plan based on that type of information will be very different from a business plan based on a bullish, high economic growth estimate. In a sense, economists can provide some useful information for planning purposes; they can't, however, project the future with certainty. This is a very similar situation to that experienced by the medical, financial planning, and legal professions. One is better off acting on informed information, even if the actual estimates of time-to-recovery, portfolio earnings, or days-to-closure for a house are a bit indefinite. Ignoring the doctor, financial planner, or lawyer does not lead to success; the same is true for economics.

In short, an off-the-mark forecast can be useful if it forces one to choose between competing scenarios and sharpens the planning effort. Whether unemployment is 5.8 or 6.2 percent is probably irrelevant from a planning point of view: A bad economy is clearly not a good economy. Even if one forecaster predicts prosperity and the other predicts recession the forecasting effort alerts the decision maker to the need for contingency planning and the determination of alternative strategies as related to economic scenarios.

This is one in a series of commentaries by the Research staff of the National Association of REALTORS®. Read more commentaries >

Comments? Questions? E-mail NAR Research.

NAR members, learn how you can add this commentary to your Web site, blog, or newsletter. Read more >