Wednesday, June 13, 2007

Dissecting Mandel

Journalist / blogger Michael Mandel of BusinessWeek, has a provocative article that claims that GDP was overstated by a rough estimate of $66 billion over the three years. In other words, real GDP should be about 0.5% lower today, than it otherwise should be. Lets review his assertions:

1. Productivity as a whole should be lower over the last few years, reported as 1.8%, and should be 1.6%
2. Manufacturing output growth is overstated by 40%

From this he draws these implications:

1. The "speed limit" for the economy is lower, implications for Fed policy
2. US is losing competitiveness
3. Current metrics are outdated

Let us look at his evidence.
"When Houseman first uncovered the problem with the numbers that is created by offshoring, she was primarily focused on manufacturing productivity, where the official stats show a 32% increase since 2000. But while some of the gains may be real, they also include unlikely productivity jumps in heavily outsourced industries"

Why, may I ask are they unlikely? Logically this makes no sense. In fact, it is intuitive that productivity would rise the fastest in the most heavily outsourced industries. Let us consider why. The first factories to be outsourced are the least competitive ones, and hence the ones with the lowest productivity if we assume wage rates are approximately the same. As an example. I have three factories, one with a productivity rate of 50, one at 100, and one at 150. Now, lets say I offshore 1/3 of my factories. I would choose the least productive. Average domestic productivity before the offshoring occurred was 100, now with the two remaining factories it is 125, an increase of 25%, due primarily to offshoring. So, it is actually rather likely productivity would jump in the offshored industries.

"Yet Washington's official statistics show that productivity per hour in the furniture industry went up by 23% and output by 3% between 2000 and 2005. Those numbers baffle longtime industry consultant Arthur Raymond of Raleigh, N.C., who has watched factory after factory close. "And we haven't pumped any money into the remaining plants," says Raymond. "How anybody can say that domestic production has stayed level is beyond me." "

Well, when you think about it, it really isn't baffling, because logic tells us that by removing poorly performing factories out of a fixed pool, by definition must raise productivity. Mr. Raymond really should take some math classes and not act so surprised, yet Mandel uses this as evidence to back up his claim of uncompetitive American workers. "
"In some sectors, productivity growth may be an indicator not of how competitive American workers are in international markets," says Houseman, "but rather of how cost-uncompetitive they are."

Time for school for you too, Mr. Mandel. Mr. Raymond needs a friend.

So we can dismiss Mandel's assertion in #2, there hasn't been a change in American competitiveness to be concerned about. Resources are being allocated as the market determines. Nothing wrong with that. The competitive factories stay, the bad ones go, and America is better off through lower prices.

I think this also begins to chip away at #1. Let us finish the demolition job, for Mandel makes another crack at supporting assertion #1 with "evidence." His "evidence":
Yet no matter how hard you look, you can't find any trace of the cost savings from offshoring in the import price statistics. The furniture industry's experience is particularly telling. Despite the surge of low-priced chairs, tables, and similar products from China, the BLS is reporting that the import price of furniture has actually risen 6.7% since 2003.

The numbers for Chinese imports as a whole are equally out of step with reality. Over the past three years, total imports have climbed by 89%, as U.S.-based companies have rushed to take advantage of the enormous cost advantages. Yet over the same period, the import price index for goods coming out of China has declined a mere 2.3%. "

Well, he has no evidence. He says import prices have risen 6.7% despite the "surge". Yet, he somehow fails to put a number on the surge. And, even further, if we adjust for inflation, which by rough calculations has risen by 8.3% since 2003 (thru 2006). His price "increase" is actually a decrease of 1.6%. Funny how that works out. So in real terms, the unenumerated "surge" has actually worked to decrease prices.

The next paragraph tells us absolutely nothing, except that China has increased its exports to the United States. This is his "evidence", on which he bases his calculation pyramid... So, color me not convinced on #1.

As for the rest of his complaints about the import price index does not take qualitative changes into account, well myself and many other economists have the exact same complaint about the CPI, and in fact health care too. So, my big response will be, yes... and your point?

#3, I agree. I think some metrics we do use are outdated and could be improved. Capital investment is a big one. But, if our objective is to describe the general welfare of people within the political borders of the US, I think we do a good job for a difficult task.

Light Posting

Sorry for the light posting, had a lot of stuff going on at work, and thats the word of it... a lot of interesting things have happened in the last month, so expect a flurry...