By Dirk van Dijk, CFA
Yesterday’s double dose of bad economic news seriously raises the possibility that GDP growth in the third quarter will be negative. That does not mean a double dip for sure, since the general rule of thumb is that a recession is two back-to-back quarters of negative GDP, although the official definition by the National Bureau of Economic Research (NBER) is a bit more complex than that.
Also, the NBER, which is the official arbiter of these things, has not yet ruled on when the Great Recession ended. Most of the signals seem to indicate that they will probably put the official end as being July 2009. It is not unusual for the NBER to rule long after the fact on recession dates. They didn’t declare the Great Recession was underway until November of 2008, roughly a year after they determined it had started.
Just as a point of clarification, a recession is when the economy is going down, and it ends when it stops going down, not when it has recovered to its pre-downturn peak. The end of a recession is the point of maximum economic pain, but the early part of a recovery is just as painful as the recession period.
To understand why we might see negative economic growth in the third quarter, it is useful to go through the components of GDP. Just as a quick reminder, GDP (Y) is the sum of spending by Consumers (C), Investment (I) and Government spending (G), and Exports (X) minus Imports (M). Thus the equation that is taught in the first day of every macroeconomics course:
Y = C + I + G + (X – M)
It therefore follows that the change in GDP, or GDP growth (dY) (I’m using a d because I don’t know how to type a triangle or delta, which is the normal math symbol for change) must be equal to the sum of the changes in the components or:
dY = dC + dI + dG + (dX –dM)
Well let’s go through each of those items, along with the major components of each. It is important to realize that they are not all the same size, and some of the components tend to be very stable, while others can change dramatically.
The largest part of GDP by far it the Consumer. In the 2Q10, it was 70.4% of GDP. For the most part, C is generally fairly stable in terms of percentage moves, but since it is such a large share of GDP, small changes have a big impact.
There are two major things that consumers their money on: goods and services. Of these, services is the much larger of the two, accounting for 67% of all Consumer spending and 47.24% of GDP. Spending of services tends to be relatively stable.
Spending on goods can be further broken down into spending on durable goods (DG) such as cars, appliances and other big-ticket, long-lasting items, and spending on non-durable goods (NDG), which are smaller-ticket and tend to be used up quickly, such as food, clothing and gasoline. Spending on NDG is about twice as large as spending on DG. However, NDG spending does not change that much with the overall state of the economy.
Durable goods spending, in contrast, is very sensitive to the overall economy. If you are worried about losing your job, or have already lost it, you are far more likely to put off buying a new car than you are to put off buying more toilet paper. Since the car was made to last for a long time, you can use it for another year when times get tough. Regardless of how bad things are economically, people are not going to reuse their toilet paper. They might trade down to a cheaper brand, but they are not going to stop using it.
Thus DG spending really “punches above its weight” when it comes to GDP growth. In the second quarter, C added a total of 1.15 points to the overall 2.39% GDP growth (using the numbers from the first release; they are going to be revised down sharply on Friday). Of that, DG contributed 0.53 points or 46.1% of the overall C contribution, even though it made up just 10.5% of overall C spending.
DG in this context is a bit different from the DG in the new orders for durable goods report we got yesterday (see Durable Goods Report: Dismal), but is related. The new orders report includes long-lasting items bought by businesses as well as by consumers. That spending is part of "I," which I will get to later.
The shrinking consumer confidence numbers probably mean that consumer spending on DG is going to be soft in the 3Q. Consumers have been hurt by a massive loss of wealth from the collapse of the housing market, and as a result are trying to rebuild their balance sheets. That means the focus is on paying off existing debts, not taking on new debts to buy more big-ticket items.
That was true in the second quarter as well, but still I suspect that spending on durable goods is not going to rise by anything like the 7.5% we saw in the second quarter. I suspect we will be lucky to see growth of more than 3.0% in the third quarter.
We will know more when we start to get the retail sales numbers for August in a few weeks. The numbers were soft in July, rising only 0.4%. Excluding autos, which is the quintessential consumer durable good, sales were up just 0.2%. Spending on NDG and services will probably grow at about the same rate as they did in the second quarter. If that were to happen, the overall positive contribution from the consumer to GDP growth would be roughly 0.85 points.
Along with DG, it is usually "I" that determines if the economy is booming or in a bust. That is not because Investment is a big part of the economy; it really is pitifully small relative to history and most other advanced economies. Aside from its vital role in supporting future growth and output, Investment is particularly important in that it is very volatile. Investment comes in a variety of “flavors.”
Overall, I (technically gross private domestic investment) has been a vital contributor to the GDP growth that we have seen over the past few quarters. In the 2Q, it added 3.14 points to growth (or more than 100% of the total growth in the economy). In the 1Q10 it added 3.04 points to growth (82.2% of total growth) and in the 4Q09 it added 2.70 points to growth (54.0% of total). That is pretty impressive for a sector that is only 12.66% of the total economy. Thus not only has Investment been a big contributor over the past several quarters, its importance has been growing.
There are several components to I. First, we need to make the distinction between fixed investment and inventory investment. Inventory investment is generally considered a very low quality source of growth. If we are making things, but they are simply being stacked on store shelves and not being sold, that growth is not going to last very long.
Big moves in inventory in one quarter tend to be reversed in the following quarter. During the financial meltdown, we had a massive inventory liquidation as businesses cut their inventories even faster than sales were falling. That was part of the reason the decline in GDP was so sharp in the 4Q of 2008 and the 1Q of 2009.
We have now had three straight quarters where inventory rebuilding was a major contributor to GDP growth. In the 2Q10 it added 1.05 points to growth, down from adding 2.64 growth points on the 1Q and 2.83 points in the 4Q09. In other words, inventories have already been rebuilt, and it would be foolish to think that we will get any help on that front in the 3Q.
If we are lucky, change in inventories will simply not be a factor either way, but we might even have an inventory drawdown, which would be a subtraction from GDP growth. Let’s just assume no contribution either way from inventories
Fixed investment added 2.09 points to growth in the 2Q. There are two parts to fixed investment: residential (RI, mostly home building) and non-residential (NRI, business investment). The homebuyer tax credit actually caused RI to contribute 0.59 points to growth in the 2Q. That came on the heels of a very long string of quarters where RI had been a major drag on economic growth. However, the tax credit really did not stimulate any new demand for housing -- it mostly changed the timing of those who would have bought anyway.
RI is a very small part of the overall economy, especially now after so many quarters of big declines. At this point it only makes up 2.46% of GDP. That such a small sector of the whole economy can generate so much growth is a testament to just how volatile RI can be.
RI was up 27.9% in the 2Q (although I suspect that the growth rate and thus the contribution to growth will be revised down tomorrow). Based on the data we have already seen on housing starts, building permits and new home sales, not only is RI not going to repeat in its contributing to growth, it is almost assuredly going to once again be a drag on growth in the 3Q. It is still early, and we need the data for August and September, but I would pencil in about a 0.3 point drag to growth in the 3Q, not a 0.59 point addition.
NRI has two major components: investment in equipment and software (E&S) and investment in structures. Investment in structures actually surprisingly added to GDP slightly (0.14 points) in the 2Q after being a major drag in the previous two quarters. Like residential investment, this is a very small part of overall GDP (2.65%), but it really swings around and thus its health has a lot to do with how well the economy overall is doing.
Right now, vacancy rates are very high for almost all forms of commercial real estate. Thus there is not a huge incentive on the part of business to put up more office buildings, strip malls or hotels. It would not shock me to see the small positive contribution from structure investment revised away tomorrow. The American Institute of Architects (AIA) billing index has been consistently in negative territory (i.e. below 50 as it is one of those magic 50 indexes) since January 2008.
There's a three- to four-quarter lag between architects sending out their bills and actual construction spending (which is what counts for GDP). Based on where the index was late last year, it seems very likely that structure investment is going to be falling in the 3Q, and thus will be subtracting from GDP growth. As for how much of a drag, it sure would not surprise me if it was roughly as much as in the 1Q, when it subtracted 0.53 points from growth. Let’s just call it a 0.50 point drag.
Equipment and software investment has been one of the real strong points in the current recovery, adding 1.36 points in the 2Q on top of contributions of 1.24 points in the 1Q and 0.91 points in the 4Q09. That is a pretty nice-size help from a sector that makes up just 6.97% of the whole economy. Unfortunately, it seems likely that the contribution will be much smaller in the 3Q than it was in the 2Q.
The best proxy for E&S investment comes from the New Orders for Durable Goods report in the form of what is known as "core capital" goods. That is, non-defense capital goods, excluding aircraft. After rising sharply in the first half of the year, core capital goods fell by 8.0% in July. While July is only one third of the quarter, clearly E&S investment is not getting off to a strong start.
While the forecast might change significantly if the core capital goods numbers improve in August and September, right now it seems a good idea to assume that the contribution from E&S investment in the 3Q is less than half what it was in the 2Q. Let’s call it 0.60 points.
Government spending (not including transfer payments; changes in transfer payments like Social Security show up in C as Grandma spends her check) is 20.51% of GDP at all levels. In the 2Q, government as a whole added 0.88 points to growth, with 0.72 coming from the Federal government and 0.16 points coming from State and Local governments.
We are close to the peak of spending related to the ARRA or Stimulus Bill, and that spending will start to wind down in the second half and into 2011. While the spending in the 3Q will help hold up the level of GDP, since the spending will be less in the 3Q than in the 2Q, the tailing off of stimulus spending will actually become a drag on GDP growth.
It strikes me as likely that while the contribution from the Federal government will be positive in the 3Q (the ARRA is not the only thing the Federal government does), that the contribution will be much smaller, perhaps adding just 0.3 points to growth (most of which will probably come from the defense-spending side).
State and Local
State and Local governments also were net contributors to GDP growth in the 2Q, adding 0.16 points. This was also a bit of a surprise, and it would not be shocking to see that revised away as well tomorrow.
In any case, for most of state and local government entities their fiscal years start in the beginning of July. Most are projecting very large budget deficits, but are not allowed by law to run operating deficits. To the extent they balance their budgets by cutting spending, this will show up as a drag in GDP growth from S&L governments in the 3Q. If they balance them by raising taxes, it will tend to show up as lower growth in consumer spending.
I would pencil in a 0.50 point drag in growth from the S&L government side. That means as a whole, Government will turn from being a 0.88 point contributor in the 2Q, to a 0.20 point drag in the 3Q. If anything, I am probably being too generous -- the drag could well be much more than that.
We don’t have any data yet on net exports in the 3Q. The last data we have is for June, but that was just plain ugly, with the trade deficit rising from $42.0 billion in May to 49.9 billion in June. The trade deficit has been in a steep upward trend in recent months, and it was a huge drag on growth in the 2Q, subtracting 2.78 points, versus subtracting 0.31 points in the first quarter.
Actually, the drag was much larger than that. The 2Q figures I am referencing are from the first look, and were issued before the June trade deficit numbers were available. They were much worse than expected, and trade is going to be the single biggest reason for the big downward revision in 2Q GDP growth we are going to see tomorrow.
If the trade deficit simply holds at the June level in July, August and September, the drag from trade will be at least as big in the 3Q as the 2.78 point drag we currently see in the 2Q numbers. Let’s just round to a 2.80 point drag.
So let’s add it all up and see what sort of growth we are looking at for the third quarter:
Perhaps I was too pessimistic in some areas. There is still a lot of data that has to come in, and that data could change the assessment in a big way. However, that being said, it is not difficult to see how GDP growth could easily be negative and fairly significantly so in the third quarter.
If I had to guess, I would say I might have been a bit harsh on the Consumer front, and perhaps the Trade data will not be quite that awful. Still, it is looking more and more like when the 3Q GDP data comes out, the headline number will have a minus sign in front of it.