he always said he'd never do that.  Last night though it sounded like he was having second thoughts about that.  Shocked me.

But what do they know?  They just study this stuff for a living....

Since last spring, when fears of economic apocalypse began to ebb, we've been treated to an alphabet soup of predictions about the recovery. Various economists have suggested that it might look like a V (a strong and rapid rebound), a U (slower), a W (reflecting the possibility of a double-dip recession), or, most alarming, an L  (no recovery in demand or jobs for years: a lost decade). This summer, with all the good letters already taken, the former labor secretary Robert Reich wrote on his blog that the recovery might actually be shaped like an X (the imagery is elusive, but Reich's argument was that there can be no recovery until we find an entirely new model of economic growth).

No one knows what shape the recovery will take. The economy grew at an annual rate of 2.2 percent in the third quarter of last year, the first increase since the second quarter of 2008. If economic growth continues to pick up, substantial job growth will eventually follow. But there are many reasons to doubt the durability of the economic turnaround, and the speed with which jobs will return.

Historically, financial crises have spawned long periods of economic malaise, and this crisis, so far, has been true to form. Despite the bailouts, many banks' balance sheets remain weak; more than 140 banks failed in 2009. As a result, banks have kept lending standards tight, frustrating the efforts of small businesses--which have accounted for almost half of all job losses--to invest or rehire. Exports seem unlikely to provide much of a boost; although China, India, Brazil, and some other emerging markets are growing quickly again, Europe and Japan--both major markets for U.S. exports--remain weak. And in any case, exports make up only about 13 percent of total U.S. production; even if they were to grow quickly, the impact would be muted.

Most recessions end when people start spending again, but for the foreseeable future, U.S. consumer demand is unlikely to propel strong economic growth. As of November, one in seven mortgages was delinquent, up from one in 10 a year earlier. As many as one in four houses may now be underwater, and the ratio of household debt to GDP, about 65 percent in the mid-1990s, is roughly 100 percent today. It is not merely animal spirits that are keeping people from spending freely (though those spirits are dour). Heavy debt and large losses of wealth have forced spending onto a lower path.

So what is the engine that will pull the U.S. back onto a strong growth path? That turns out to be a hard question. The New York Times columnist Paul Krugman, who fears a lost decade, said in a lecture at the London School of Economics last summer that he has "no idea" how the economy could quickly return to strong, sustainable growth. Mark Zandi, the chief economist at Moody's Economy.com, told the Associated Press last fall, "I think the unemployment rate will be permanently higher, or at least higher for the foreseeable future. The collective psyche has changed as a result of what we've been through. And we're going to be different as a result."

One big reason that the economy stabilized last summer and fall is the stimulus; the Congressional Budget Office estimates that without the stimulus, growth would have been anywhere from 1.2 to 3.2 percentage points lower in the third quarter of 2009. The stimulus will continue to trickle into the economy for the next couple of years, but as a concentrated force, it's largely spent. Christina Romer, the chair of President Obama's Council of Economic Advisers, said last fall, "By mid-2010, fiscal stimulus will likely be contributing little to further growth," adding that she didn't expect unemployment to fall significantly until 2011. That prediction has since been echoed, more or less, by the Federal Reserve and Goldman Sachs.

The economy now sits in a hole more than 10 million jobs deep--that's the number required to get back to 5 percent unemployment, the rate we had before the recession started, and one that's been more or less typical for a generation. And because the population is growing and new people are continually coming onto the job market, we need to produce roughly 1.5 million new jobs a year--about 125,000 a month--just to keep from sinking deeper.

Even if the economy were to immediately begin producing 600,000 jobs a month--more than double the pace of the mid-to-late 1990s, when job growth was strong--it would take roughly two years to dig ourselves out of the hole we're in. The economy could add jobs that fast, or even faster--job growth is theoretically limited only by labor supply, and a lot more labor is sitting idle today than usual. But the U.S. hasn't seen that pace of sustained employment growth in more than 30 years. And given the particulars of this recession, matching idle workers with new jobs--even once economic growth picks up--seems likely to be a particularly slow and challenging process.

The construction and finance industries, bloated by a decade-long housing bubble, are unlikely to regain their former share of the economy, and as a result many out-of-work finance professionals and construction workers won't be able to simply pick up where they left off when growth returns--they'll need to retrain and find new careers. (For different reasons, the same might be said of many media professionals and auto workers.) And even within industries that are likely to bounce back smartly, temporary layoffs have generally given way to the permanent elimination of jobs, the result of workplace restructuring. Manufacturing jobs have of course been moving overseas for decades, and still are; but recently, the outsourcing of much white-collar work has become possible. Companies that have cut domestic payrolls to the bone in this recession may choose to rebuild them in Shanghai, Guangzhou, or Bangalore, accelerating off-shoring decisions that otherwise might have occurred over many years.