In a much-anticipated speech on Friday, Fed Chairman Ben Bernanke invoked a favorite metaphor of economists to describe the current, critical period in the recovery. Now is the moment, he said, that a “handoff” must occur between temporary boosts to growth, like government stimulus, and more lasting drivers, like spending by consumers and businesses. Bernanke stressed that the handoff was underway, but he conceded that “growth has been proceeding at a pace that is less vigorous than we would like.”
The particular source of his anxiety was consumer spending, which the chairman worried was flagging amid high unemployment, sideways home prices, and the debt we all took home from the recent subprime party. It comes as no surprise that consumers would weigh on the Fed chairman, of course. While it’s certainly possible for the economy to grow without consumers pitching in, it’s much, much easier when they do. Because consumer spending accounts for about 70 percent of GDP, a 2 percent increase raises GDP growth by almost 1.5 percentage points. By contrast, it would take a roughly 10 percent increase in investment—spending on things like factories, equipment, and new-home construction—to boost GDP growth by roughly the same amount, because investment only accounts for about 14 percent of GDP.
Unfortunately, even under relatively optimistic scenarios, consumer spending isn’t likely to impress anybody over the next several years. Instead, the backdrop for Bernanke’s comments is a debate about whether spending by consumers will fall only modestly relative to the pre-crash days, or whether we’ll see a pronounced drop that weighs on the economy’s back like a large, belligerent primate. Here’s hoping for mediocrity.
It’s not hard to find proponents of the more upbeat scenario (such as it is) among private-sector economists. But the Obama administration has made the case for it with more rigor and precision than just about anyone around. The administration’s model, laid out in the Economic Report of the President in February, assumes that three basic factors drive the decision to save rather than spend: wealth, unemployment expectations, and access to credit. During the boom years, the housing market boosted wealth, jobs were plentiful, and credit flowed easily (which is of course the definition of a credit bubble). Not surprisingly, Americans saved little and spent briskly.
Once the bubble burst, this process went into reverse. The saving rate—the fraction of income people choose to save—steadily rose from about 1-2 percent before the recession to around 6 percent in the first half of this year. A rising saving rate means consumption is growing more slowly than income. Put differently, consumers have been a drag on the economy.
Here’s where the tentative optimism comes in: While unemployment is likely to stay uncomfortably high for the foreseeable future, wealth is gradually recovering (as the stock market rises) and banks are extending more credit. The combination of those things should stabilize the saving rate at 6 percent, or even bring it down a bit. That means consumer spending should grow at about the rate that incomes grow, or slightly better, going forward. “We were in a period in which people had to ratchet down their level of spending,” says Chris Carroll, an economist at Johns Hopkins University, who drafted the White House report on consumer spending while at the Council of Economic Advisers. “We may be seeing the end of the period where it has to be ratcheted down.”
Of course, the bad news is that incomes aren’t likely to grow very impressively over the next few years, so consumer spending won’t either. “In my view there’s not much reason to think that households are going to power us out of current slow period,” says Carroll. But at least they won’t be exacerbating the problem.
The beauty of Carroll’s model is that it explains, with uncanny precision, consumer behavior going all the way back to the late ’60s. Those three simple variables—wealth, unemployment, and credit—tell you most of what you need to know about changes in the saving rate, and their predictive power has held up even through 2010. (Indeed, when they were compiling the president’s report, Carroll and his fellow White House economists were initially alarmed to find that the saving rate had been much lower of late than the model predicted. Then, the official government data-collectors at the Commerce Department released one of their periodic revisions, and the new numbers turned up exactly where Carroll had expected them to be.)
The question—really more like a nagging terror—is whether something has happened since the recent financial crisis to fundamentally change the way consumers behave, rendering the administration’s model moot. As it happens, there’s a school of wonks that worries this is the case. The godfather of this group is a Japanese economist named Richard Koo, whose framework for thinking about this appears in a book he modestly titled The Holy Grail of Macroeconomics. (Paul Krugman, among others, has identified himself with some of Koo’s ideas.)
Koo’s view is that consumers and businesses who take on enormous debt during a bubble abruptly shift gears once the bubble bursts, spending very little while they pay off loans. Moreover, this stinginess continues until the process of debt-repayment (economists call it “deleveraging”) is complete, creating a huge drain on the economy. In the case of Japan, whose real estate and stock markets collapsed in the early ’90s, this took over a decade. During that time, Koo argues, the only force propping up the economy was massive amounts of government stimulus. He tells a similar story about the Great Depression.
Whereas Carroll assumes people base their saving decisions on the same factors both before and after the crisis, Koo says the way they make decisions beforehand tells you little about their behavior afterward. The crash doesn’t just pummel the value of their assets (like housing). It creates a kind of psychological trauma that preoccupies them with paying down debt before they can think about borrowing again. If you accept Koo’s premise, the data of the last 40 years is of little help in guiding us through the current situation. The episodes we’re talking about—Koo calls them “balance-sheet recessions”—simply didn’t happen at any point in that time-frame.
So how can we tell who’s right? It turns out both approaches make similar predictions about what should have happened since the financial crisis—both forecasted that saving would shoot up. It’s only now that their predictions are diverging. Carroll believes the saving rate depends most heavily on wealth and credit availability. (Unemployment has historically been a less important factor, though its importance could increase during a deep recession for “precautionary” reasons, in economist-speak.) Koo believes leverage (i.e., the amount of debt you have relative to your income or assets) is what really wags the dog. In Koo’s view, access to credit is almost meaningless. It doesn’t matter how much banks are willing to lend if you’re so obsessed with paying down debt you can’t contemplate another loan.
Carroll is somewhat optimistic because wealth (at least relative to income) is about where it stood between the mid-’80s and mid-’90s, and well above where it was the decade before that. He also notes that banks are gradually becoming more willing to lend. By contrast, adherents of Koo’s view would stress that, pretty much any way you measure leverage, it’s receded only slightly from the historical high it reached after the recent boom*—implying that it would take years before consumers are ready to spend freely again.
Just how long are we talking about? Early last year, economists at the San Francisco Fed observed that, if you extrapolate from the Japanese experience, the deleveraging process would take about a decade, during which time the saving rate would rise to about 10 percent, subtracting about half a percentage point from GDP growth each year (a huge amount when GDP is only growing by 2-3 percent).** Slightly less alarmingly, the economist Allen Sinai has constructed an index of household financial conditions based on the measures of leverage we’re talking about. Sinai says the index recorded its all-time worst reading in early 2009 and estimates it’ll take another two or three years to get back to a level that’s healthy by historical standards.
We should be able to figure out whether we’re living in Chris Carroll’s world or Richard Koo’s over the next few six to nine months; the first big set of indicators—data on spending and saving from this year’s third quarter—should be out in the next few weeks. In either case, the economy probably needs more stimulus—9.6 percent unemployment is much too high by any measure. But if it’s Koo who better approximates reality, the stimulus need could be acute at a time when GOP congressional gains have made it a political nonstarter. As I say, mediocrity has never looked so attractive.
Noam Scheiber is a senior editor of The New Republic and a Schwartz Fellow at The New America Foundation.
*Some have pointed out that, while the ratio of debt to income or debt to assets remains very high, debt-servicing costs (basically interest payments) aren’t so out of whack with their historical averages, thanks to extremely low interest rates. That is, we may have a lot of debt, but it’s not costing us a ton to keep up with. There’s something to this, but interest payments can be misleading. As economist Jan Hatzius of Goldman Sachs pointed out in a September research note (not online), once you adjust for inflation, debt-service costs are actually very high by historical standards.
**Of course, the saving rate doesn’t necessarily have to keep rising in order for households to pay off debt. Households can do so when they save 6 percent of their income; it just takes longer than if they’re saving 10 percent. The assumption is just that households will want to pay it off sooner rather than later, hence the rising saving rate.