Wednesday, January 27, 2010

The Ugly Truth Behind Friday's GDP -- Gross Domestic Pain

from Minyanville:

On Friday, the Bureau of Economic Analysis will release its first estimate of Gross Domestic Product (GDP) for the fourth quarter of 2009. It’s widely expected that the number will be a positive 4% to 5%. That will be the second consecutive quarter of positive GDP growth, which “officially” means that the recession is over.

The talking heads will greet this number with great enthusiasm. They will interview members of the administration, which will greet the number with even greater enthusiasm proclaiming that a long period of steady economic growth is right ahead of us, all of which is due to its economic policies and actions.

Yeah, right.

In advance of this stellar achievement, I thought it might be useful to look at how the two most important economic indicators, GDP and the unemployment rate, are calculated. I have to admit I forgot the definition of Gross Domestic Product (GDP) from my undergraduate economics class, so if you also forgot, or skipped class that day, here it is:

GDP = C + I + G + (X-M)

C = Private consumer consumption, basically all the goods and services you buy

I = Investment, which includes business investment of plant, equipment, inventory, structures, etc. It doesn’t include investment in financial assets (your day trade that you accidentally held overnight does not contribute to GDP).

G = Government spending, all salaries, military, and investment, but no transfer payments, such as social security or unemployment benefits.

X = Exports

M = Imports

You might have noticed that there’s no consideration of debt, either personal or government, in the calculation of GDP. So, if the government spends like crazy (which it is), that spending will boost GDP.

China’s fourth quarter GDP came in at 10.7% -- to pull that off you need to do stimulus right. There’s no better example than this video. Every brick in that empty city is pure GDP. That stimulus makes our Cash for Clunkers and road repaving look pretty trivial.

With stimulus in mind, let’s look at our third-quarter 2009 GDP numbers.

GDP increased 2.2%, but most of that increase, 1.45%, was due to increased automotive purchases, obviously due to the Cash for Clunkers program. So a “G” program drove an increase in “C” without a subtraction of the increased national debt that was created by Cash for Clunkers. For actual numbers, click here.

Table 3 shows the actual numbers in billions of dollars. The quarterly numbers are seasonally at annual rates and the headline reported numbers are from the chained 2005 dollar column, which adjusts them for inflation. Look at two lines (shown here) in the Personal Goods section, motor vehicles and recreational goods and vehicles (SUVs fall into this category).


Click to enlarge


In the second quarter, total sales in these two categories fell $9.7 billion from the first quarter. In the third quarter, they increased by $45.4 billion from the second quarter. Because that's an annualized rate, the clunker program generated an additional $11 billion in sales at a cost of $3 billion. Whether or not you believe it was a good program, it did work and produced a pretty good multiple on the cost.

Total GDP increased $71.5 billion in the third quarter, but Personal Consumption only increased $63.6. Subtracting Clunkers leaves only $18.2 billion. Scanning all the other goods and services categories shows, on average, small gains and losses, the biggest of which was food at $6.6 billion -- not exactly an area where consumers enjoy spending more money. Overall, I'd hardly call it the start of the end of the recession for consumers.

The problem with GDP is that it doesn’t represent what the average American family is feeling. Much closer to consumer sentiment is the unemployment rate. You might think that the GDP calculation -- trying to get all spending and investment from all sectors -- would be difficult, but in reality, GDP has three months to get the final revision in, while the unemployment rate has little more than two weeks to be guessed at.The biggest guess part of the unemployment rate is called the birth/death model. The concept is that the Bureau of Labor Statistics (BLS) can't possibly contact all the small businesses that are starting, closing, hiring, or firing during the one-week period that the unemployment rate is measured. The birth/death model tries to account for these unmeasured jobs by guessing each month the number of jobs created or lost in several sectors of the economy. For the birth/death model, click here.


Click to enlarge


As you can see, the birth/death model created jobs every month in 2009 except January. During that time frame, the unemployment rate went from 7.4% at the end of 2008 to 10.0% at the end of 2009. If you find that small businesses managed to create net jobs all those months hard to believe, welcome to the club.

Despite the birth/death model, deciding who's working and who isn't is very complicated for the BLS. The “headline” unemployment rate (currently 10.0%) is calculated as follows: The number of unemployed divided by the labor pool.

That looks simple enough, but there's a lot going on under the surface. If you get laid off from a middle-management job and are working 20 hours a week at a fast-food joint to make ends meet, but you're still looking for a full-time job, you're fully employed according to the headline number.

There's one more item to discuss. If you're unemployed and give up looking for a job, you're now a “discouraged worker” and you're dropped from the labor pool. Here's a simple example to show the effect of discouraged workers.

Let’s say a small town has 100 people in the labor pool and 90 of them have jobs. The employment rate is 90/100, which is 90%, so the unemployment is 100%-90% = 10%. One of the unemployed workers gives up and stops looking for work. 90 are still working, but the labor pool is now 99 people. The employment rate is now 90/99 which is 90.9% which yields an unemployment rate of 100%-90.9 = 9.1%.

Amazing! If enough people stopped looking for work, we could get the unemployment rate way down and turn the economy right around! Mark Twain popularized the phrase : “There are three kinds of lies: lies, damned lies, and statistics.” I think Twain would have really liked the headline unemployment rate.

To be fair to the BLS, it does publish additional unemployment rates that can be found here.

The broadest group of underemployed workers (including discouraged and part-timers looking for full time) is reported by the BLS under the enlightening term, U6. U6 was 13.5% at the end of 2008, and ranged between 17.0% and 17.4% for the last four months of 2009. May I suggest a slightly more descriptive name for U6. How about Gross Domestic Pain? Isn’t this the real GDP that the average consumer is feeling?

GDP will rocket this Friday, but there was no improvement in the Pain Index in the fourth quarter. The combination of stimulus and inventory building will produce a good GDP number that will be unsustainable without progress in getting people back to work. With talk that Obama wants to cap discretionary spending, I don’t see how a new stimulus package can be seriously considered until the economy is in free fall again, and considering the deficit, an inability to finance a new stimulus package is a real possibility.

In sum, don’t get fooled by the strong GDP number on Friday. The real GDP, the Pain Index, will control the economy. I've raised cash, expecting to see a double-dip recession that I don’t believe the market has started to discount. I plan to trade SDS as a hedge to my long-term positions, but I emphasize trade. For example, I don’t expect to be hedged Friday morning before the GDP number, but I would expect a good fade opportunity soon after. That’s my plan and I’m sticking to it.