10 August 2009 – Chart of the Week



Chart of the Week: The Personal Savings Rate gets revised up again, and again, and again ….

Until recently, if it seemed that the headlines were almost always highlighting a falling savings rate, that’s because the data as it’s initially reported often shows a falling savings rate, which is subsequently revised up sharply in later years. The savings rate, which basically measures what everybody earns minus what they spend, has some inherent flaws. For instance capital gains in the stock market are not added to savings but capital gains taxes are subtracted from income. Even so, the trends can be striking, at least at first blush.

For each monthly release, the government’s statisticians typically revise the numbers for the last month or two by relatively modest amounts. However, every few years they scrub the data from top to bottom; these benchmark revisions can be substantial and they are almost always upward, which was again the case in the latest round of revisions that were recently released. Fortunately, the Philadelphia Fed preserves the entire history of the vintage data at their website.

In 1980, for instance, the data as it was then reported showed a precipitous decline in the savings rate from around 7% to just over 3%. By 1986, that decline was revised away but it was replaced by a new decline to below 3%. The savings rate even turned negative a few times and much ink was spilled on the parlous condition of Americans’ savings, first in 1999 and then again in 2006, both of which were subsequently revised away. By last year, the savings rate appeared to be about flat and in this year’s benchmark revisions the savings rate was lifted for most of the past decade to around 3%, with a recent recession-driven surge to 5.2%.

The persistence of upward revisions is usually due to the discovery of greater income growth than was originally calculated rather than less spending. An undercounting of income in the underground economy might also help explain a secular decline in the average savings rate that appears in the 1980s, although that trend is arguably also due to the advent of alternative channels for savings, notably 401ks. Or it is entirely possible that ten years from now it will all have been airbrushed away in future benchmark revisions. The history of the savings rate provides an excellent case-study in the limits of economic data and the need for healthy skepticism about the models that use that data as well as the claims that follow.



The information presented in this presentation (the “Presentation”) has been prepared solely for informational purposes and is not an offer to buy or sell or a solicitation of an offer to buy or sell Interests or any other security or instrument or to participate in any trading strategy. No representation or warranty (express or implied) is made or can be given with respect to the accuracy or completeness of the information in this Presentation. No person has been authorized to make any representations concerning the Interests described in this Presentation that are inconsistent with the statements contained in this Presentation.