GDP Growth: Accelerating or Decelerating?

For most of the week, the U.S. stock market continued to seesaw, sometimes violently, but ended down a mere 0.4% (S&P 500) for the week. Even relatively strong GDP and consumption data for December didn't really help the market. The stabilization of emerging-market stock prices, which were virtually unchanged for the week, didn't put the U.S. markets in a better mood, either.

A report out of Europe on Friday suggested that inflation in Europe was a mere 0.7% recently, raising fears of deflation and Japanese-style malaise. Bonds also had another decent week, with 10-year U.S. Treasury bonds now yielding 2.67% after starting the year at close to 3%.

Not helping the market at all this week were corporate earnings, which continued to be mixed at best.  Wal-Mart (WMT) was one of the companies that provided lower earnings guidance, noting the impact of a smaller food stamp program as well as certain emerging-markets issues. At the other end of the retailing spectrum, e-retailer  Amazon (AMZN) also posted results that fell below expectations. The hit parade of earnings losers on Friday alone included  Mattel (MAT),  Newmont Mining (NEM), and  MasterCard (MA). The news wasn't all bad, though, with nice surprises out of  Chipotle Mexican Grill (CMG),  Computer Sciences (CSC), and Zynga .

The economic news was also very mixed. Durable-goods orders, sometimes a good predictor for the manufacturing sector, looked terrible. Pending home sales also managed to find yet another cliff to fall off of, dropping almost 9% in December. Home prices seemed to hold their own, and weekly shopping center data looked a teensy bit better for the week.

The GDP data was good at 3.2% growth for the fourth quarter, although there are some questions of sustainability and special one-time helps. Consumption in December, with 0.2% inflation-adjusted growth, was better than I thought. However, it appears that strong utility and winter garment sales may have provided some of that boost. Unfortunately, the extra spending came out of savings, and real disposable incomes dropped by 0.2% between December and November.

I continue to believe that the economy is chugging ahead at its same 2.0%-2.5% pace that it has for some time. Although there were some modest signs of acceleration in some reports this week, I think weather factors may have provided a short-term boost (apparel, natural gas, groceries, gasoline) that will quickly turn into negative factors in the months ahead. Poor income growth won't help much, either. Housing continues to have issues, but it is very hard to separate out the weather effects from a genuine slowing shown in the past two weeks of data.

Fourth-Quarter GDP Report Shows Decent But Lopsided Growth
The overall fourth-quarter GDP growth rate met expectations at 3.2%, though not quite the way we all expected. The consumer did well, growing at 3.3%, but that was below expectations of 4% growth. Residential investment shrank a surprising 9.8%, as falling commissions on existing-home sales hammered this category hard. Government also did far worse than expected, shrinking almost 5%.

Offsetting these negatives was a far larger contribution from net exports than almost everyone expected. The number was almost in the too-good-to-be-true category. The number is especially suspect given the results for the last month of the quarter have to be estimated until actual export data is released next week. Recall also that November net exports were surprisingly strong with a massive drop in imports and modest export growth. Larger-than-expected inventory growth also offset other weaknesses in the report. Inventories had made a huge jump in the third quarter, and most folks were surprised that inventories grew at an even faster rate in the fourth quarter. At some point this rather large inventory build will need to reverse itself, negatively affecting GDP growth in future quarters.

The table below shows the growth rate for the various categories in addition to which percentage of GDP the category represents. The final contribution column is simply the growth rate multiplied by the percentage of GDP. The sum of the contribution column equals the total growth in GDP for the quarter. Notice how that, even though the services category showed one of the lower growth rates, the large size of this category made it the second-largest contributor to GDP growth.

GDP Growth Accelerating or Decelerating?
The exact period one picks makes a relatively large difference in guessing what the economy's true trend is. Full-year-to-full-year data shows relatively strong decelerations, the sequential-quarter data shows good growth but a decelerating trend, and the year-over-year single-quarter data shows good, accelerating growth.

Although it is the most widely cited, the sequential quarter-over-quarter data has never been a favorite of mine, and I believe it distorts economic reality. Note how volatile the middle column of that table is. It doesn't mesh well with employment and consumption data, which has been far more stable.

There are really two major problems with this data point. First, it takes very small quarterly changes and multiplies them by four to get to an annualized rate. Any estimation errors, which can be large, are magnified in this regime. Also, large and sometimes shifting seasonal factors become a much bigger factor when comparing sequential quarters with varying seasonal factors than comparing two quarters a year apart that usually have only modestly different seasonal factors. The full-year-over-full-year data of the last column fully strips out those seasonal factors, allowing the true data to shine through. Unfortunately, this metric suffers from a timeliness issue, because it includes four quarters of data and misses the fact that early 2013 was quite slow and growth showed at least some acceleration in the second half while just the reverse was true in 2012.

My usual preferred measure is the full-quarter-over-the-full-quarter data displayed in the first column, which shows a healthy 2.7% growth rate and an accelerating trend. But this might be just a little too optimistic this time around, as 2012's fourth quarter was slowed by Hurricane Sandy, and this year had the export anomaly and abnormally high purchases of anything related to cold weather. Still, I think this is the metric closest to the mark, and I suspect the current underlying growth rate is somewhere in the 2.0%-2.5% range for the fourth quarter and showing steady improvement.

Consumer Spending Is Up, but Incomes Aren't Keeping Pace
Over the third quarter, income growth (as measured by real disposable income) exceeded consumption growth in each individual month of the quarter. That situation exactly reversed itself in the fourth quarter when consumption growth far exceeded income growth in every month, as shown below.

The size of the gap and the period that it has lasted both seem out of line with recent history, suggesting that in the first quarter, the United States will suffer another bout of slow consumption and, hopefully, better incomes. Although most economists were excited about consumption growth during the fourth quarter (which was the second-best result of the whole recovery at 3.3%), I remain at least a little suspicious. I fear that high utility bills (which boosted GDP) might have helped inflate the fourth quarter but may haunt the economy in 2014, as the bills for that energy are actually paid. Sales of winter weather apparel and stocking up on groceries and gasoline in the face of oncoming storms all probably gave the consumption data a little boost in December.

Pending home sales continued to fall in December, hitting their lowest level since 2011. Although higher rates and stronger home prices have plagued this index since May, this month was likely affected by poor weather conditions. But even averaging the data and looking year over year, the trend in pending home sales has been abysmal.

Year-over-year data showed an outright decline in both November and December. Though weather certainly played some role, sales in the West, where weather has been unusually good, showed one of the biggest sequential declines, while the hard-hit Midwest showed the smallest decline. So it's not all about the harsh winter, either. And January may prove even worse than December, as conditions worsened substantially.

Pending home sales are important because they are a leading indicator of closed existing-home sales. It is those closed deals that generate remodeling, moving, furniture buying, and mortgage brokerage activity that is helpful to the economy. Gauging by the gap between the pending sales index and the existing-home sales index, existing-home sales are likely to fall even more in the early months of 2014.

Housing Price Indexes Continue to Grow but No Longer Accelerating
This week Case-Shiller released the last of the three major home price indexes for November. The three-month moving average crept up slightly to 13.8, trending similarly to the  CoreLogic data but better than the FHFA database that is showing slower growth each month. Sunbelt markets again led the way in November, with many other markets languishing: Las Vegas, Los Angeles, and San Francisco all grew faster than 20% year over year. Meanwhile, New York, Washington, and Cleveland all grew 8% or less.

That will also mean rough going for the housing investment portion of the GDP calculation in the first quarter, although perhaps not as bad as 0.3% negative contribution in the fourth quarter. All of that decline was due to falling brokerage commissions on existing-home sales while both remodeling and new home construction were virtually unchanged in the fourth quarter. Since home prices bottomed in March 2012, prices have increased by almost 24%, according to the Case-Shiller 20-City Index.

Given that the preponderance of last year's housing gains were in the first half, those year-over-year gains could remain quite elevated for the next several months before dropping to the 5% range by the end of 2014. (Although it's clear to many that the quick and easy gains in real estate pricing may be over for the time being.) Current low mortgage rates--significantly lower than just two months ago--might provide some modest help to home prices, if the weather could just give the economy a break.

Durable Goods Orders: Weather Issues or Warning Sign of Impending Weakness?
On the surface, the durable goods report looked bleak. Overall orders were down 4.3% from November to December, largely because of falling airline orders and softer auto orders. Still, excluding the volatile transportation sector, orders were still down 1.6% in a month that should have benefited from a rush to take advantage of expiring tax credits and alleged renewed business confidence. Though weather could have had some impact, poor weather is more likely to affect shipments than orders.

Because of the volatility in the overall data set, plus the fact that airline orders make little difference in the short- or even intermediate-term production results because  Boeing (BA) has a decade's worth of backlog already, this report has lost a lot of its meaningfulness and predictive ability. I do like to look at the nondefense, capital goods orders excluding aircraft as an indicator of business confidence, more than anything else. Here the news isn't quite as bleak but the trend is still not positive.

The reason this is mainly just an indicator of confidence, rather than a great forecaster of the manufacturing sector, is that this subsector represents just 29% of total durable goods.

A by-category analysis of total durable goods orders doesn't shed a great deal of light on the situation. Of the six major categories, only machinery orders and the tiny electrical equipment sector showed any order growth on a month-to-month basis. Computers and electronics orders continued to struggle, as they have all year, with orders down 7.8% month to month and down 2.8% year to date. Every other major category was up on a year-to-date compared with a year-ago basis, suggesting things aren't quite as soft as the month-to-month numbers imply. Besides weather, strong and changing seasonal adjustment factors may have skewed this month's report to the downside.

Manufacturing, Auto, Employment Data, and Trade Data on Deck for Next Week
The ISM data for manufacturing is due on Monday, and expectations are for a small decline from 56.5 to 56.0. Given weather, emerging markets, and this week's poor durable goods report, a softer number wouldn't come as a big surprise to me. Again, consumption data is more important than manufacturing data at this stage of a recovery.

Auto sales are also due on Monday, and expectations are for a decent rebound from a poor December. Some sales that may have normally occurred in December could have drifted into January. Sales are expected to move from a relatively disappointing 15.4 million units in December to 15.8 million units in January. However, I still worry that weather could have put a dent in sales. There were many days during the month when even a test drive would have proven difficult across wide sections of the country. Auto sales have been a key to this recovery, and the data remains a little sloppy and very volatile.

Payrolls grew a shockingly low 74,000 in December, perhaps because of weather or seasonal factors. The trend had previously been about 190,000 new jobs per month. Analysts are expecting a move back to that trend line in January with job growth of 189,000.

Trade was a big help to the recent fourth-quarter GDP report, so economists will be watching to see how closely the government estimated net exports for December. A big variance would mean a big GDP revision. The deficit was $34 billion in November, the lowest level of the year. Expectations are for a slightly higher deficit at $36 billion. I expect higher imports and less-robust growth in oil exports, which could lead to a more normal deficit level of $39 billion.