The figure below shows a series of forecasts by the Federal Reserve for real GDP growth in 2015, with the first forecast made in March 2013 and the last one made a few days ago. Each bar shows the central range of the forecasts by Fed economists, with the average forecast as a dot in the middle of each range.
The punchline, as you see, is that the closer we’ve gotten to the actual outcome for GDP growth this year, the more they’ve marked down their previous guesstimates. Back in early 2013, they thought we’d be cooking along at 3.3%, and if anything, their “north-of-three” forecast became stronger as time went on, as seen by the narrower range around the average.
Then, as reality set in and faster growth turned out not be around the next corner, their forecasts came down a bit. Still, they thought ample growth rates just shy of 3% were awaiting us this year.
Except they weren’t, and the markdowns continued and got bigger, with the most recent one taking their 2015 guesstimate from 2.5% down to 1.9%. As the NYT crisply put it, “In a retreat that has become a ritual for the overly optimistic central bank, officials said in a new round of economic forecasts published Wednesday that they expected the economy to grow this year by 1.8 percent to 2 percent.”
Now, to be clear, this isn’t meant to pick on the Fed forecasters, who are no better or worse than most others. You can see this same pattern in the forecasts of the IMF, the World Bank, CBO, and the National Association of Business Economists.
What’s interesting here are the two questions this raises: 1) what’s driving these markdowns? What’s going on in the economy that top forecasters keep systematically missing, and 2) Why aren’t the forecasts self-correcting? Why do they keep trusting Lucy to hold onto the football?
Re the first question, there are numerous perps that have contributed to the markdown pattern.
–Bad fiscal policy: While we’re not Europe, we too pivoted to deficit reduction too soon. The figure below, from Goldman Sachs researchers, shows the impact of fiscal policy on GDP growth, 2009-15. In 2009, the Recovery Act did some heavy lifting, but it quickly faded (despite President Obama’s efforts to do more) and in 2013, fiscal policy was a big negative on growth and jobs. The fiscal growth impulse has shifted into neutral for now, and I suppose “do no harm” is the most we can expect from the current Congress. But there’s no question that our subpar growth has been a function of unlearning lessons about the need for supportive fiscal policy in weak economies.
–Weak investment: Both public and private investment has been subpar in recent years, with the latter feeding back negatively into productivity growth, as Larry Mishel shows—quite dramatically—here. In terms of our transportation infrastructure, according to the most recent World Economic Forum rankings, the U.S. fell from 7th to 18th in the quality of our roads over the past decade, as our investments in this space have declined by half as a share of GDP since the 1960s. I should note here that the Republican budget resolution calls for a 40% decline in transportation funding.
–Deleveraging and the reverse wealth effect: I’ve written in lots of places how debt bubbles, like those involving mortgages, take a lot longer to work through then equity bubbles. Basically, your shares in Pet Rocks, Inc., get marked to market quickly, while banks holding non-performing loans can “extend-and-pretend” ad nauseam. Interestingly, Chair Yellen has often made this point herself, referring to this dynamic as a persistent and ongoing headwind. To be fair—and this gets a bit into question #2 (why are the models systematically wrong)—no one really knows the timing of how this kind of dynamic plays out in the macroeconomy.
–Inequality and the decline in the compensation share of national income: Economist Joe Stiglitz has argued that the extent of income inequality has also hurt growth in ways the models underweight. If you’re just thinking about averages, you may miss the fact that while average income has gone up, median income has gone up a lot less—this is a symptom of inequality. In the same vein, the share of national income going to paychecks versus profits is around a 50-year low.
The germane factor here re growth is that those with higher propensities to spend the marginal dollar are seeing fewer dollars flow their way, while those who are anything but liquidity constrained—the very wealthy—are bathing in the stuff. As I note below, the housing bubble and its wealth effect offset this dynamic in the 2000s. But not now.
–Congressional dysfunction: IMHO, this is a bigger deal than people think. It’s de rigueur to say stuff like, “well, as long as Congress is tied up in ideological knots, they can’t mess stuff up.” But, in fact, that’s just wrong. First, businesses can’t plan ahead if they don’t know what’s happening with tax issues like bonus depreciation and other such “extenders.” Second, as a colleague who watches both governments and markets pointed out to me, the return on government investments is lower when it’s made at the last minute, without adequate planning. Again, think of highway and mass transit spending. And of course, debt ceilings and shutdowns don’t help, either.
Re the second question—why haven’t the forecasters self-corrected?—it is by definition the case that the models are inadequately capturing the factors above, including any I’ve left off. However, before you trash the forecasting endeavor in general, and you wouldn’t be far off to do so, consider that it’s awfully tricky to build Tea Party obstructionism into an economic model. Also, the impact of inequality on consumption growth is well-established theoretically, but less so empirically, in part due to the wealth effect noted above. Thus, there are complex interactions beyond the scope of current models.
End of the day, my advice is: don’t think of these point estimates as reliable predictions. Instead, think of them as a) what the Fed thinks, which has implications for monetary policy, and b) more importantly, as a clear message that if we want to get to and stay at full employment, we’ll need to reverse the negative practices listed above. That is, we’ll need better fiscal policy, investment in productive infrastructure, less inequality, and functional politics.
Gosh, when you put it that way, it doesn’t sound so hard, right?!