Are The Rich Responsible for the Low US Savings Rate?

George Will writes:

Americans' saving habits are better than they seem because the very rich, consuming more than their current earnings, have a negative savings rate.

That really surprised me as I thought that the rich were responsible for more of the saving in the economy.

I am not sure what data George Will was looking at, but Google Answers pointed me to this Federal Reserve document and in 2000 it was true that those with the highest 20% of income had a -2.1 savings rate while all the lower quintiles had positive savings rates.

But, as the Federal Reserve document explains, this was only due to the fact that stocks had been greatly increasing leading up to 2000 and those in the upper 20% saw their investments were doing so well that they were taking extra spending out of their assets. Even while they were selling some of their assets, the market was doing so well that their total asset value was increasing.

If you go back to 1992, when investments weren't doing as well, the top 20% actually had the highest savings rate of any quintile at 8.5%. My guess is that with the current downturn in the economy that the rich will be leading the way in savings once again.

As I was looking into this, I found out that there is more than one way to define savings and more than one agency that reports on it. The Bureau of Economic Analysis releases the Personal Savings Rate number. But, this has some problems as My Money Blog points out:

The personal savings rate ignores the capital gains in our investments like stocks and bonds, and also tangible assets like cars and real estate.

Seeking Alpha gives a great example of how that can lead to a misleading result.

The Federal Reserve releases the Flow of Funds Accounts which is another way to look at savings. The FFA number does take into account realized capital gains and counts the net acquisition of consumer durable goods as savings. I am not sure how it handles real estate though.

This document explains the difference between the two methods. I would have thought then that the FFA method would lead to a higher savings rate, but if you look at page 4 in this document, there are a couple of times where FFA leads to a lower savings rate.

The numbers I described above used the FFA method, and shouldn't have the issues brought up by Seeking Alpha and My Money Blog.

As for whether the rich are responsible for the low US savings rate, I think they were when during the late 90s when there was a stock market bubble and also in the recent real estate bubble. They were spending more than they were earning because their investments were doing so well. But, now that the bubble has burst, expect that the rich will be leading the way in savings once again.

Update: Richard Cooper points out another form of savings that is overlooked by these measurements: education.