The latest GDP numbers, that supposedly showed that real GDP expanded 0.6% in the first quarter of 2008 after having expanded at the exact same rate the previous month, was used by a number of economists, mainly supply-side Republican ones, to deny that America is in a recession. First and foremost among them was Larry Kudlow, along with Jerry Bowyer, whose book The Bush Boom Kudlow wrote the introduction of. The argument made is that since GDP supposedly still expands and since recession is supposedly defined as two consecutive quarters of falling GDP, this means that there aren’t no recession.
But as a matter of fact, both parts of this argument is wrong. First of all, recession is in fact usually not defined by two consecutive quarters of falling GDP. Although there were three quarters of falling GDP in 2000 and 2001, these declines were interrupted by quarters of rising GDP, so using the definition of two consecutive quarters of falling GDP there weren’t any recession in 2001. Instead, recession is usually defined as a broad based decline in economic activity, which specifically means falling industrial production, payroll employment, real disposable income excluding transfer payments and real business sales. This, together with some indicator of monthly GDP is how the National Bureau of Economic Research (NBER) define a recession.
And although monthly GDP estimates aren’t available to us mere mortals, the other 4 indicators are available to us, and they are summarized by the Conference Board in its index of coincident indicators. And the current numbers for the index of coincident indicators shows that the recession is indeed here, and more specifically began in November 2007. The index has fallen each month since October 2007, except for March 2008. And that number looks likely to be revised down. Previous months numbers have systematically been revised down (Two months ago, they claimed that the coincident index in January 2008 were 0.2% higher in January 2008 compared to October 2007. Now they say the index fell 0.2% in that three month period) and considering the fact that two of the components, real disposable income and real business sales, were a mere imputation due to lack of actual data. Since then, we have gotten the number for real disposable income in March which did not rise 0.25% as the Conference Board assumed, but instead was unchanged.
But it’s not only the index of coincident indicators which tells us that the recession began late last year. In fact, using a more proper price index to deflate the increase in nominal GDP gives us the same conclusion and does in fact show two consecutive quarters of falling GDP. If we use a price index that underestimate inflation, this will not only lead us to underestimate price inflation, but it will also given a certain level of nominal GDP growth lead us to overestimate real GDP growth.
Many readers will at this point probably think that I will argue that the government price indexes underestimate price inflation and that some alternative measure, from for example John Williams’ Shadow statistics web site is better. However, while there is a case for believing that the price inflation numbers produced by the government underestimate price inflation ( their methodology have been revised numerous times, curiously always resulting in lower inflation), my argument that real GDP has fallen for two consecutive quarter does not depend on doubting the validity of government price index. Even if we accept for the sake of the argument that the government price indexes are completely accurate, that still means that real GDP has fallen for two consecutive quarters.
Instead, my case rests on pointing out that the headline GDP number is derived by deflating the nominal GDP increase by the wrong government price index. That means specifically, the Gross Domestic Product price index instead of the Gross Domestic Purchases. The first index is based on the prices Americans sell, while the second index is based on the prices Americans buy. To a large extent, these price indexes overlap as they both include the prices of things Americans sell to other Americans. The difference is that the Gross Domestic Product index include the price of exports while excluding the price of imports while the Gross Domestic Purchases index exclude the price of exports while including the price of imports.
When export- and import prices increase at the same rate, these two price indexes will increase at the same rate. But when import prices rise faster than export prices, then the Purchases index will rise faster and conversely when export prices rise faster than import prices the Product index rill rise faster.
As the point of GDP statistics usually is said to be to indicate how much purchasing power for Americans their production will give them, it makes much more sense to deflate nominal GDP growth with the price index of what they buy rather than the price index of what they sell. If the prices of the things you buy fall, then other things being equal this will increase your real income. But if the prices of the things you sell fall, then this will other things being equal lower your real income.
Your real income certainly don’t rise if you work 2% more but your hourly wage fall 3% and the prices of the things you buy are unchanged. But according to the methodology used to produce the alleged positive GDP number, real income in that case did in fact rise 2%. Similarly, the net income of a company who cuts the price of its product by 20 % while increasing volume sales by 10% at a time when the prices of the things the owners and workers of that company are unchanged. Even assuming no fixed costs, that would of course imply a 12% decline in real profits and real wages (For simplicity I assume that wages fall too. The basic point does not however rest on this assumption).But according to the methodology used to produce the alleged positive GDP number, we are supposed to assume that the real income for the owners and workers rose 10%.
And what is absurd for individual workers and companies is also absurd for aggregates of workers and companies, like the GDP number.
Having established that it is more proper to deflate nominal GDP with the Gross Domestic Purchases index, we can now re-evaluate the GDP numbers. Nominal GDP growth was 3.0% in Q4 2007 and 3.2% in Q1 2008, yet the price index that measure the purchasing power of Americans, the gross domestic purchases deflator, rose 3.7% and 3.5% respectively, meaning that real GDP fell 0.7% in Q4 2007 and 0.3% in Q1 2008.
Illustrating the fact that this is a more proper approach, we can note that the alleged positive growth number was in part based on a alleged increase in net exports. Yet the actual trade deficit rose in fact from $708.9 billion to $737.3 billion at an annual rate. Rising net exports means a falling, not a rising, trade deficit yet the methodology used to provide the alleged positive GDP numbers still counted this increased trade deficit as rising net exports because the increase in imports were mostly based on rising import prices.
In short, there can be little doubt that America is in a recession now and that this recession probably began in November 2007. The supply-siders that deny this must resort to using the absurd logic that it is irrelevant for your living standard if you must pay more for the things you buy, or get paid less for the things you sell. Somehow, I doubt that they are willing to apply this logic in their personal life and accept a lower pay for their propaganda pieces.
May 5, 2008