There are two different stories people tend to tell when they're trying to explain why the middle class is feeling squeezed.
The first one is about inequality. The top 0.1 percent of earners -- that's one taxpayer out of every 1,000 -- now brings in 11 percent of the nation's total income, triple the share that they did just a generation ago. This has happened because the rich have grown ever richer, while the pay of rank-and-file workers hasn't risen much faster than inflation.
The second, related story is about instability. Layoffs seem to happen more frequently than they once did, and these job losses -- combined with the spread of bonus pay -- have caused workers' incomes to bounce around a lot more than in the past. So not only have middle-class families been getting meager raises, their finances have also become more volatile.
The story about inequality is indisputably true. But we're starting to learn that the second story, the one about instability, is more complicated. It may even end up being wrong.
There is now a big push in both Washington and state capitals to come up with policies that can alleviate middle-class anxiety. That's all for the good. In fact, it is overdue. If it's going to succeed, however, it will have to focus on the actual causes of the squeeze.
Last week, the Congressional Budget Office released a study that was arguably the fullest picture of economic volatility anyone has yet put together. Although some academics have taken a crack at the topic in recent years, they have had to rely on surveys in which people are asked how much money they make. The study by the C.B.O., as the budget office is known, used Social Security Administration records, which cover many more people than the surveys and are more reliable.
If you read the C.B.O. report, you can tell that its authors knew they were dealing with a delicate subject. The summary starts by noting that a ''significant number of workers experience substantial variability in their total wage earnings,'' which is certainly true. Only later do you come to the surprising part: there is the same amount of variability now that there was in the 1980s and 1990s. In journalism, this is known as burying the lead.
''Intuitively, you would think volatility is increasing,'' said Senator Charles E. Schumer, Democrat of New York, who along with Senator Jim Webb of Virginia requested that the study be done. ''But it isn't, which I guess shows that the American economy has always been very flexible.''
Mr. Schumer's point about intuition is an important one. We can all tick off reasons that the economy feels so volatile. Hardly a week goes by without another big corporation -- the Tribune Company, Citigroup, DaimlerChrysler -- announcing a big job cut. The number of temporary jobs, meanwhile, has mushroomed. Globalization and technological innovation are causing many of these changes, and labor unions are too weak to prevent them.
But there is also a whole set of other forces, harder to see and pushing in the other direction. Manufacturing, where furloughs and layoffs have always been the norm, accounts for a much smaller part of the work force than it used to, while more stable industries, like health care, have grown. This is one reason that recessions, and the job cuts they bring, haven't happened as often as they once did.
Yesterday's layoffs could go unnoticed by people who weren't affected by them because they were a regular part of doing business. Today's tend to come in big chunks and are often announced in news releases. ''There have always been a lot of mass layoffs,'' said Lawrence F. Katz, a labor economist at Harvard. ''We didn't count them before.''
In fact, research by Henry S. Farber, an economist at Princeton, has found that job loss rates have followed a cyclical pattern since the early '80s, peaking around the same highs during recessions and falling to similar lows during expansions. (The rate has risen for workers who went to college and fallen a bit who those who didn't.)
Americans, looking at their own jobs, realize that there hasn't been a big change: in a recent Gallup Poll, 12 percent of respondents said it was very or fairly likely they would be laid off in the coming year. In the 1970s, '80s and '90s, at similar points in the business cycle, the percentage was virtually identical.
I don't want to exaggerate how much we know about instability, because there are some conflicting signals. Mr. Farber has also found, for instance, that average job tenure has declined during this time. Maybe people are more willing to leave a job voluntarily -- or maybe companies nudge more workers out the door with early retirements and other stealth layoffs.
The C.B.O., for its part, has looked only at the pay of individual workers. In the future, economists there will look at family income -- which is affected by divorce and other factors -- and could potentially find changes there, as some academic work has.
The bottom line, though, is that the picture is likely to look mixed. Volatility may or may not have increased over the last generation, but it does not appear to have changed in a fundamental way.
Inequality is a whole different story. It has risen enormously and is not about to stop. With the recent jump in gas prices, the pay of rank-and-file workers is once again failing to keep pace with inflation, just as it did from 2002 until early last year, despite healthy economic growth.
In fact, inequality is probably the real reason that the economy often feels more volatile. When people are stretched -- when their pay has been stagnant, when they're worried about health insurance, when they don't know what the future holds -- a jolt to their income is harder to handle.
As the C.B.O. pointed out, these jolts do happen a lot, even if they don't happen more than in the past. In 2003, a whopping 20 percent of workers saw their earnings drop 25 percent or more compared to the previous year. (For about 22 percent of workers, earnings rose at least 25 percent.)
But it's important to keep in mind what is really changing. As Mr. Schumer says, ''If you're holding a job but your share of the pie is getting smaller, that's a different set of policy needs than if you keep losing your job.''
In an economy where volatility was the main problem, you might want to protect jobs by making it harder for companies to cut them. In an economy where inequality was the problem, you would want to protect people. You would help them pay for health insurance, retirement, their children's education and other basic needs when the market, left to its own devices, was not doing so.
And if your resources were limited, wouldn't you start with the problem you were sure that you had?