GDP growth in the second quarter was revised up to 1.7% from a second read of 1.6% but still down from the preliminary reading of 2.4%. This post is essentially a re-post of my analysis of the second reading.
The current estimates are in bold, and the previous two estimates are inside the parenthesis, with the second estimate in italics, and the first look at the numbers in regular type. This will allow you to see just where the changes have come from in the revision, as well as how the second quarter stacks up relative to earlier quarters. New commentary, or changes in directional words (i.e. rose or fell) are also in bold.
Very Little Change
In general, there was very little change from the second reading. The bump up of 0.1% came from the lowest quality source of growth, more inventories. That was offset by an even worse deterioration in the trade picture than previously estimates. The consumer was notably stronger then in the first estimate.
The increased drag from net exports is very serious, and needs to be addressed. A weaker dollar would help, but the thing that would help the most is if the country got serious about ending its addiction to imported oil. However, I expect that Vladimir and Estragon will be having a long, leisurely chat with Godot before that happens, given the current political climate.
The bottom line on getting serious about ending our oil addition is putting a price on carbon emissions, either through a direct tax or through a more indirect cap and trade system. The reduced contribution from inventory investment is not serious. If it had not happened in the second quarter, it would have happened in the third quarter.
In the second quarter (2Q) Gross Domestic Product grew at 1.7% (1.6, 2.4%), which was slightly better (less) than the 1.6% (1.4, 2.5%) consensus expectation. That is sharply slower than the 3.7% growth rate of the first quarter.
Since different parts of the economy are of very different sizes, but some of the biggest parts tend to be relatively stable and some of the smaller parts can change dramatically from quarter to quarter, in this post we will concentrate on how much each sector contributed to (or subtracted from) GDP in terms of growth points. The points will sum to the 1.7% (1.6%, 2.4%) overall growth in the economy.
A Clearer Picture
I think that gives a clearer picture of what is happening in the economy rather than focusing on the percentage change in each sector. When I do mention percentage changes, they are at annual rates. I will follow the familiar Y = C +I + G + (X- M) framework, where GDP is the sum of consumption plus investment plus government plus net exports.
Without a doubt, growth has been slowing from the 5.0% rate in the fourth quarter to 3.7% in the first quarter to the current estimate of 1.7% (1.6, 2.4%) growth in the second quarter. However, the quality of the growth has been improving. In the second quarter, 0.82 (0.63,1.05) points of growth came from the change in inventories (non-fixed investment), down from 2.64 points in the first quarter and 2.83 points in the fourth quarter.
The biggest part of the economy by far is Consumption, accounting for 70.4% of total GDP in the 2Q. In the 2Q, Personal Consumption Expenditures (PCE) added 1.54 (1.38, 1.15) points of growth, down from 1.33 points in the 1Q but up from the 0.69 point addition in the 4Q. People spend on both goods and services, and goods are further broken down into durable goods, such as cars and furniture, and non durable goods like food and clothing (although the government clearly has not looked in my closet if they consider clothing to be non-durable, I think I still have some stuff that dates from the Reagan Administration).
In total, goods added 0.79 (0.82, 0.79) points to growth in the 2Q, down from 1.29 points in the 1Q but up from 0.42 points in the 4Q. Of that, durable goods added 0.49 (0.49, 0.53) points versus 0.62 points in the 1Q, but up from a 0.07 point subtraction from growth in the 4Q.
Spending on durable goods was 7.36% of the entire economy in the 2Q. By the very nature of being durable, spending on durable goods tends to be easy to postpone when times get tight. Instead of going out and buying a new car from Ford (F) or Toyota (TM) when people are worried that they might get laid off in the near future, they simply drive the old clunker a little longer.
That demand, then, gets pent up (you get tired of driving that old thing or the repair bills start to mount) and when good times return, the spending on durable goods tends to jump. Thus, durable goods tend to “punch above their weight” when it comes to determining if the economy is in a recession or is booming.
Spending on non-durable goods is a much bigger part of the economy at 15.77%. However, non-durable goods spending tends to be much more stable than spending on durable goods. In the 2Q, non-durable goods consumption added 0.31 (0.33, 0.25) points to growth versus 0.67 points in the 1Q and 0.49 points in the 4Q.
Most of consumer spending, though, is not on stuff — it is on services, accounting for 47.24% of the whole economy. Despite its huge size, services added just 0.75 (0.56, 0.36) in the 2Q up from just 0.03 points in the 1Q and 0.27 points in the 4Q. Services tend to be performed in real time, and can not be stored, thus they tend to be a more stable part of the economy.
The part of the economy that really punches above its weight is investment. Gross Private Domestic Investment (GDPI) is really the thing that makes the difference between boom and bust. It is broken down in to fixed and non-fixed investment, with non-fixed being the change in inventories I discussed above.
Inventory investment is considered low-quality growth, since if factories are making things that are simply piling up on store shelves, it means that they will have to cut back production in the future. Fixed investment, on the other hand, is a bet on the future of the country, and for the most part adds to the country’s productive capacity.
In total, GDPI makes up just 12.66% of the overall economy, but it was responsible for 2.88 (2.75, 3.14) points of growth in the 2Q, down from 3.04 points in the 1Q and 2.70 points in the 4Q. The change in fixed investment is even more dramatic, with fixed investment adding 2.06 (2.12, 2.09) points in the 2Q up from 0.39 points in the 1Q and an actual subtraction of 0.12 points in the 4Q.
Fixed investment is further broken down into residential (homebuilding and improvements) and non-residential. Non-residential investment was 9.62% of the economy in the 2Q and added 1.51 (1.54, 1.50) points to growth, up from 0.71 points in the 1Q and a subtraction of 0.10 points in the 4Q. It is further broken down into investment in structures, such as the building of new office buildings and shopping centers, and into investment in equipment and software (E&S).
Spending on structures subtracted 0.01 (+0.01, +0.14) points to growth, but that is a big positive swing from the subtraction of 0.53 points in the 1Q and a 1.10 point subtraction in the 4Q. That improvement is a major surprise, and I would not expect it to last (it was revised almost completely away). There are simply too many vacant office buildings and empty stores in the country for it to make sense to be building a lot of new ones.
This now makes it the eighth straight quarter that investment in non-residential structures has been a drag on growth. Spending on non-residential structures has fallen by 33.7% over the last two years, and in the process it has declined from 4.04% of the whole economy to just 2.65% of the economy.
Investment in E&S added 1.52 (1.53, 1.36) points to growth, up from 1.24 points in the 1Q and 0.91 points in the 4Q. E&S spending increased by 24.9% (21.9%), and now represents 6.97% of the whole economy, up from 6.43% a year ago. This is a very encouraging development, suggesting that businesses are starting to deploy some of the massive cash hoard they have amassed.
The sharp increase in E&S spending is all the more surprising since manufacturing capacity utilization is only at 72.2%, which is an extremely depressed level — the long-term average is 79.2%. If businesses are investing even when they have factories sitting idle, it means that they must be getting more confident about the future.
On the other hand, it is partly a reflection of the sharp decline in E&S spending that happened during the recession. Even with the sharp increase, E&S spending is still 8.7% below where it was two years ago. From the recent report on new orders for durable goods, it looks like the momentum in E&S investment has slowed, and in the third quarter the contribution it makes to overall GDP growth will be smaller.
The other side of fixed investment is residential investment (RI). It added 0.55 (0.58, 0.59) points to growth, but I would not expect that to last. Mostly it is due to the homebuyer tax credit, which drew demand into the 2Q from the 3Q and 4Q. RI jumped at an annual rate of 27.2% (27.9%) in the 2Q. However, there is a huge overhang of existing homes for sale, especially when one counts the shadow inventory of homes where the owners are far behind on their mortgages and are likely to go into foreclosure (or which are already in the foreclosure process).
RI has been a perpetual thorn in the side of the economy, having subtracted from GDP growth in 13 of the previous 14 quarters. Since the peak of the housing bubble, RI has fallen from 6.43% of GDP down to just 2.46% in the 2Q, even with the big (and artificial) jump in the 2Q.
While eventually RI should return to a more normal level of about 4.2% of the economy, I don’t think that the rise in the 2Q is the start of that process. Historically, RI has made a huge difference in determining if the economy is booming or is in a bust. Housing is in some ways the ultimate durable good, as a house will last a lot longer than a car, and used houses are even better substitutes for new houses than used cars are for new cars.
Recent data on building permits and housing starts indicate that residential investment will again be a drag on growth in the 3Q, and quite possibly 4Q as well. The just-plain-awful new homes sales numbers for July and August also point in that direction.
Government spending was responsible for 0.80 (0.86, 0.88) points of growth in the 2Q, a big swing from the 0.32 point subtraction from growth in the 1Q and the 0.28 point drag in the 4Q. The Federal Government was responsible for 0.72 (0.72, 0.72) of that, up from a contribution of 0.15 points in the 1Q and a 0.01 point contribution in the 4Q.
Most of the swing has come from spending for defense, which added 0.40 (0.39, 0.40) points after adding just 0.02 points in the 1Q and being a 0.13 point drag in the 4Q. Non-defense spending added 0.32 (0.32, 0.33) points in the 2Q up from a contribution of 0.13 points in the 1Q and adding 0.14 points in the 4Q.
Government spending for calculating GDP is very different from the government budget, since it excludes transfer payments like Social Security. That spending is part of Consumption, and is counted when Grandma spends her Social Security check. We are close to the high point of stimulus spending, and it will soon start to tail off. While the remaining spending will help support the level of GDP, since it is falling it will actually become a drag on GDP growth from this point forward. The drag will be relatively slight in the third quarter, but will grow in significance as we move through 2011.
In total, government spending was 20.51% of the economy in the 2Q, and of that only 8.26% is Federal Government spending. Defense spending was 5.57% and non-defense federal spending was 2.70% of GDP. State and local governments added 0.08 (0.14, 0.16) points of growth in the 2Q a big swing from the 0.48 point drag in the 1Q and the 0.29 point drag in the 4Q.
State and local governments are generally not allowed to run deficits for operations (they can float bonds for capital improvements like roads and sewage systems). Since most are facing very large deficits due to falling tax revenues, they will have to cut spending sharply in the coming quarters, and will once again be a significant drag on the economy. If instead of cutting spending they raise taxes to balance their budgets, they are likely to reduce either Consumption or Investment spending (or both). State and local government spending was 12.24% of the whole economy in the 2Q.
Net exports were a huge drag on the economy in the 2Q, subtracting 3.50 (3.37, 2.78) points from growth. In the 1Q they subtracted just 0.31 points, and in the 4Q they added 1.90 points. In other words, if we had a closed economy with no imports or exports, the overall pattern of GDP growth would look very different, with growth of 5.20% (4.97, 5.18%) in the 2Q, versus growth of 3.39% in the 1Q and 3.10% growth in the 4Q.
The problem is not on the export side, which grew by 9.1% (10.3%) in the 2Q and added 1.08 (1.08, 1.22) points to growth, after adding 1.30 points in the 1Q and 2.56 points in the 4Q. The problem is on the import side, and increasing imports are a subtraction from GDP growth. After all, imports are things that we consume here, but which we don’t make here.
Imports surged 32.4% (28.8%) in the 2Q and subtracted 4.58 (4.45, 4.00) points from growth, after subtracting 1.61 points in the 1Q and 0.66 points in the 4Q. This is a reason for serious concern. It is the trade deficit that is what drives our indebtedness to foreign countries, not the budget deficit. After all, the budget deficits during WWII were far larger than the current budget deficits as a share of GDP, but when the war was over, we were by far the worlds largest creditor, rather than being the biggest debtor as we are now.
About half of our traded deficit is due to our oil addiction. We need to reduce our oil imports and have to do it soon, they are a cancer eating away at the economy. To do so, we need to both use energy more efficiently and to move to other sources of energy. We simply no longer have the oil reserves (2.1% of the world’s total in 2009) for “drill baby drill” to be a reasonable answer. Not when we are already the third largest oil producer on the planet, accounting for 9.0% of total production already (that’s more than Iran and Iraq, combined).
Oil vs. Natural Gas
The problem is that we consume 21.7% of the world’s oil. The disaster in the Gulf shows that there are substantial risks to trying to increase our production, although clearly we will need that deepwater production. The oil has been sitting there for millions of years, and is not going anywhere, so taking some time to make sure we are extracting it in the safest possible way make sense.
Fortunately, we have ample supplies of natural gas, thanks to the emerging shale gas plays. Natural gas is also cheap relative to the price of oil on a per BTU basis and it contributes far less CO2 per BTU than does oil (or coal for that matter). Oil is primarily used as a transportation fuel, but the technology does exist to run cars on compressed natural gas and is widely used outside the country.
Moving towards more use of natural gas as a transportation fuel would do wonders for improving our trade deficit, and as we do that, improve GDP growth. Revenues that now flow to Saudi Aramco would flow to companies like Chesapeake Energy (CHK).
Moving more of our freight by rail rather than truck (or rail for the long haul portion, with the containers loaded onto trucks for the last 50 miles or so) would also greatly reduce our consumption of oil and thus improve the trade deficit. Climate change is a VERY good reason to reduce our oil consumption, but it is far from the only one.
The senate seems determined not to do anything on energy and climate legislation. With the likely changes to the make up in the senate after the elections, it is almost impossible to see anything constructive on climate or energy policy happening in the next two years. In the long run, that inaction is doing huge damage to our economy.
Putting a price on carbon is the only way we are likely to make a serious dent in our oil and coal consumption. It will give the private sector a big incentive to be more energy efficient and to come up with alternative sources of energy. The car companies would sell more smaller, more fuel efficient cars and fewer big SUVs. That price would be a tax, either direct, or thorough the more convoluted cap and trade system.
Higher taxes are not a good thing in a recession or a weak and anemic recovery. However, if the revenues generated were used to lower taxes that are even more regressive or bigger drags on the economy, such as the payroll tax, it would end up being a net positive for economic growth in both the short-term and the long-term.
Overall, the report was a bit better than expected for the second quarter, but the upward revision, and the exceeding of estimates came from two sources. One was more goods filling the store shelves. That is very low quality growth.
The other source — more consumer spending on services — is a bit more encouraging. Those improvements were partially offset by the deterioration in net exports. In other words, we stocked the shelves, but did so with stuff made abroad. That really doesn’t put too many Americans to work.
Fortunately, the trade deficit in July was not as disastrous as the trade deficit in June, so it is possible that the drag on growth tram trade will be significantly smaller in third quarter. Trade can actually add to GDP growth even with a trade deficit, just as long as the deficit is shrinking.
The quality of the growth in the 2Q was much better than the quality of the growth in the 1Q or the 4Q. Much more of the growth was due to fixed investment and less came from inventory restocking. In particular, the big contribution from investment in E&S is highly encouraging. With corporate earnings soaring, there is a good chance that it will continue to be strong in the 3Q, although probably not as strong.
In the 3Q, inventories are likely to add nothing, or even be a drag on growth. The contribution from construction, both residential and non-residential, is not likely to continue and they will both probably revert to being drags in the 3Q.
Consumption is likely to add just slightly more to growth in the 3Q as it did in the 2Q, so it will not make up for the drag from the lack of inventory restocking or the drag from construction. If we can get imports under control, we might have a chance at a decent 3Q. Export growth is solid.
The recent weakening of the dollar is extremely helpful in that regard, but is not a silver bullet. Government spending is likely to add less in the 3Q than it did in the 2Q, with S&L spending likely to become a significant drag. Thus, while it doesn’t seem likely that we will go into a double-dip recession, neither are we likely to see the sort of robust growth that will rapidly bring down the still unacceptably high levels of unemployment, particularly long term unemployment.