According to a 2007 study released by The Conference Board, while the number of households with discretionary income has increased in recent years, actual discretionary spending power is concentrated among the wealthiest.

About 73 million U.S. households now have discretionary income, up from about 57 million in 2002, according to a report by The Conference Board. The percent of the U.S. population with discretionary income has increased to nearly 64 percent, up from 52 percent in 2002.

Now the flipside of this good news is that more than a third of US households had no discretionary income, and that was before the current crisis. So if these households spend less, they are presumably cutting into necessities.

And there is another problem here. The Conference Board again:

‘While the percentage of households with discretionary income has risen over the past several years, purchasing power remains concentrated in the wallets of the affluent,’ says Lynn Franco, Director of The Conference Board Consumer Research Center. ‘More than three out of every four discretionary dollars flows to householders earning $100,000 or more. And their average discretionary income is more than 2.5 times above average.’

If we apply this knowledge to the savings rate issue, we can generate two different, although not necessarily mutually exclusive, scenarios. One, households across the board are spending less, and the ones at the bottom of hurting because of it. Two, the increase in savings is concentrated in the wealthiest households, who have the bulk of the discretionary income and therefore the most choice of whether to spend or save.

To complicate the matter further, the definition of the savings rate is strangely narrow. Here is John Steele Gordon, on the American Heritage blog, defending Americans against charges of wild spending:

The trouble here is not with Americans; it is with the definition of the savings rate, which is hopelessly out of date. That definition is the percentage of after-tax income that is not spent. If a family takes home $50,000 and spends $47,500, its savings rate is 5 percent. If it takes home $50,000 and spends $50,500, its savings rate is -1 percent.

That definition was not a bad one in 1932 and 1933, when few American families owned their own homes, were the beneficiaries of company pension plans, held any substantial financial assets, or paid any income taxes. Today it is a meaningless statistic.

Just consider. Every time a family sends a mortgage check to the bank, part of that money increases their equity in the house. But that doesn’t count as savings. Contributions, by employer and employee, to a 401(k) or other retirement plan don’t count because they are made with pre-tax income. Unrealized capital gains in stocks, bonds, or real estate don’t count. Social Security taxes don’t count either, even though they amount to 12.5 percent of total income from salaries or wages up to a little less than $100,000.

And then Steele offers some useful context:

It might be noted that the savings rate peaked in 1977, when it was over 11 percent, and has been in near continuous decline ever since. Why? In 1978 the 401(k) revolution began, and more and more Americans, more than happy to have Uncle Sam help out, began saving out of pre-tax income rather than after-tax income. So the statistic called the personal savings rate declined while the amount of wealth being saved in the real world began to increase sharply.

And here is another defense, from late 2006—a time when the savings rate was in the basement, but banks were “bulging at the seams with record amounts of cash socked away by everyday consumers.” Michael Giusti, of Bankrate.com, explains these two seemingly contradictory facts by bringing the idea of wealth into the equation:

By many measures, even with a falling savings rate, U.S. consumers are wealthier than they have ever been.

How wealthy? Well in 2006,

Federal Deposit Insurance Corp., or FDIC, records show that American banks have more cash in their vaults than at any other point in recent history—with $6.4 trillion deposited in the domestic offices of U.S. banks as of June.

Elinore Longobardi is a Fellow and staff writer of The Audit, the business-press section of Columbia Journalism Review.