Flexible Forecasting: Looking for the Next Economic Model

Last autumn I gave a talk in California's San Fernando Valley. I was the last of three economists speaking that day, and I watched the other economists? presentations, each a rosy forecast of recovery and imminent prosperity. So, I was a bit nervous when it was my turn to speak, because I had a forecast of extended malaise. People don?t like to hear bad news, and they do blame the messenger. In the end, I was relieved. No tomatoes, no catcalls.

That?s how things went last fall and winter. Many economists confidently predicted a rapid recovery, while my group?s forecasts were pretty dismal: weak economic growth with little if any job creation. Today, many of those same economists? forecasts are far closer to ours. Why?

Part of the problem is the fact that macroeconomics is an unsettled discipline. We have lots of macroeconomic models, none of which is adequate for all states of the world all the time. Each provides insight, but no single model can cope with the awesome complexity of the world. A large part of the art of forecasting is determining which model is most applicable to the current situation; which ones include insights that are dominant today.

The problem is exacerbated when economists become excessively committed to a particular model. This isn?t religion or politics, it's forecasting. It is hard enough. There is no reason to handicap yourself by excessive fealty to some model or doctrine.

There was another problem that resulted in the change of tune. The world changed in September 2008. We call it a regime shift. It's a move from one (good) equilibrium to another (bad) equilibrium. Statistical models that worked well in the old regime don?t work in the new regime. We hustled to adjust our models, but admitted that with limited experience in the new regime, we were less confident in our forecasts.

The problem with a regime shift is that it is similar to a change in the rules of a game. Old relationships don?t hold anymore. Football is an example: If you changed the rules to allow five downs instead of four, nobody would predict punts on fourth down.

Some economists didn?t recognize the regime shift. They went about their business using the same old models in a new world. Comments about the length of a typical recession or about how sharp declines are followed by rapid recoveries were clear signals that the speaker didn?t understand the situation.

Some economists were fooled by the stimulus. The rules of accounting cause government spending to be reflected as an increase in economic activity. Stimulus plans such as Cash for Clunkers and tax credits for home purchases moved the timing of transactions, artificially reinforcing the direct spending impacts. Similarly, bailouts and foreclosure prevention programs postponed the recognition of losses.

Many interpreted the resulting increase in last winter?s reported activity as permanent, but that could not be. We were not building anything or laying the groundwork for sustained prosperity. Instead, we were just continuing the previous decade?s consumption binge. The banks had failed, but the government had stepped in. It became the mother of all banks, borrowing from future citizens and other countries to fuel today?s consumption.

Regime shifts that lead to a bad equilibrium appear to be similar to bank runs. There need be no basis for panic, but a panic can guarantee the demise of a bank. The result of a panic on a bank ends there. The bank is failed, gone. There may or may not be a contagion effect on another bank.

A panic can also guarantee an economic decline. But our economy is different than a bank. It can?t fail, in the sense that we can?t shut it down and walk away. We?re all still here after a regime shift. We?re stuck with a mess.

We did have a mess after September 2008. All of a sudden, everyone?s wealth had declined, a lot. Businesses, consumers and governments were over-leveraged. Risk aversion had increased, perhaps to remain high for decades. Our understanding of economic risks had changed. We had discovered black swans ? rare and unexpected outliers ? in our system.

The problem with regime shifts is that we don?t know how to initiate or cause them. We see shifts to bad regimes, and we can see their self-fulfilling nature. Can there be some self-fulfilling process that leads to a shift to a better regime? I hoped so, and I hoped that Obama?s election would initiate such an event. Our forecasts aren?t based on hope though, and it?s just as well that we didn?t forecast that his election would generate a spontaneous recovery.

Today, enough time has passed that even the most slowly adapting forecasters are forced to confront the post-2008 data and the government?s failed economic efforts. As forecasters confront these facts, their forecasts are becoming increasingly gloomy. Now, forecasts of protracted malaise or even a double-dip recession are increasingly common. Why?

Because we borrowed to extend a consumption binge, and we compounded that error with omissions and perverse policy.

The stimulus?s omissions are glaring. We didn?t significantly invest in infrastructure that would improve our future growth. We failed to address the weaknesses in our education sector that fuel increasing inequality, sentence many to a life of hopelessness, and permanently constrain our economic growth. We did nothing to encourage small business?s growth; in an example of perverse policy, we are actually creating a new regulatory regime that favors large companies.

Then there were the actions that will probably restrain future economic growth. The minimum wage was raised. We had health care reform, but we didn?t address the real problem: the fact that the health care consumer pays an insignificant portion of the bill at the time of consumption. We had financial reform that failed to address the fundamental problems of too-big-to-fail, and we protected risky activities, increasing the regulatory burden and crippling the ability of small banks. We halted much of our offshore drilling.

Looking forward, there is little reason for optimism. We?re considering huge increases in our energy costs through greenhouse gas regulation. We have a massive tax increase scheduled at the end of the year.

While a double-dip recession is not the most likely outcome, we can?t reject the possibility. More likely, we face a long slow struggle to overcome ourselves and restore real prosperity. The forecasters? consensus appears to be moving toward accepting that reality.

Flickr photo of Petra's Yoga Poses Around The World

Bill Watkins is a professor at California Lutheran University and runs the Center for Economic Research and Forecasting, which can be found at clucerf.org.