No-Confidence Votes for the Consumer

I have been grousing about soft consumer confidence (as measured by weekly retail sales, auto sales, and housing starts) even as the economic backdrop continues to improve (better employment, a higher stock market, and falling gasoline prices). This week, businesses in general and a major trade organization checked in with their own votes of no-confidence.

Nondefense capital goods--ex-aircraft orders--declined for the fourth month in a row. Year-over-year data is softening again and is likely to grow worse in the months ahead. Capital goods are deferrable but provide long-term growth potential and are indicative of business attitudes regarding the overall economy.

Perhaps the report from the National Association of Realtors is even more troubling. The association reiterated this month that there is likely to be no growth--zero, nada--in existing-home sales in 2014 compared with 2013. This is a trade organization that is charitably known as being on the too-optimistic side of the curve (less favorably called cheerleaders by others).

I am just guessing, but the direct effects of the government shutdown, the potential for another bruising budget battle in January, uncertainty regarding health insurance, and doubt about Fed policy are all contributing to the malaise. Some of these factors may worsen and some will potentially improve in the months ahead. What is clear is that if the fog ever lifts, businesses and consumers have the wherewithal to spend more than they are today. Equally true: If they remain scared, fear alone could cause another slump. Let's all hope that the fog lifts soon.

The economic data this week was relatively thin and terribly hard to interpret. Confusing a lot of my normally accurate year-to-year metrics was a slump last September and October due to Hurricane Sandy, which makes for easy comparisons that get remarkably hard in November and December. Then, this year had some effects from the government shutdown first slowing growth, then providing a small rebound when workers returned.

Overall, the manufacturing sector looked a little better to me as year-over-year orders and shipments of durable goods continued to improve. However, the capital goods section of the report clearly showed that businesses remain fearful about the longer term.

Turning to the real estate market, pending home sales continued in a funk, showing a year-over-year one-month decline, the first time that's happened in more than a year. Home price growth also continued to decelerate, with growth increasingly concentrated along the West Coast and in Florida. Month-to-month new home permits data looked great at first glance, topping the 1-million mark and notching its best level of the recovery. However, almost all that growth was from artificially reduced levels (caused by the spike of mortgage rates that helped sales of existing homes at the expense of slower-to-close new-home sales). The year-over-year permits data continued to slow.

Durable Goods Orders Slipping Again, but Manufacturing Sector Still OK
Durable goods used to be an exceptionally useful predictor of both manufacturing and overall economic activity. Some of that usefulness has faded as volatile airline orders, and to a lesser degree autos, are what really drive the aggregate order data, but not production data.  Boeing (BA) and auto industry production schedules are what drive hiring and economic activity, not new orders. Boeing's backlogs are so huge (extending up to 10 years) that production could continue to move upward without another new order. Also complicating the data is a lot of month-to-month volatility. Bad months follow really great months and vice versa, in many cases.

This month's data is a perfect example. Orders had been up sharply for several months (including aircraft and autos) and then fell back sharply (month-to-month 0.5%, 4.1% and negative 2.0% for August, September, and October). The year-over-year data below probably provides a truer picture of the state of the overall manufacturing industry, especially the shipments component. The new-order data is still benefiting from some particularly large airline orders from this summer, distorting even the year-over-year data. Also, the year-over-year data is currently benefiting from Hurricane Sandy-related slowdowns a year ago (which affected both September and October). Still, I think the data below paints a picture of a relatively stable, perhaps slowly improving manufacturing sector.

Slowing Nondefense Capital Goods Orders Show Lack of Business Confidence
I also like to look at nondefense capital goods orders less aircraft as a gauge of business confidence. Capital goods are an interesting measure of businesses' long-term outlooks because they last multiple years and can often be deferred.

Although this is a great way to measure confidence and can be a useful forward-looking indicator, I do caution that it is really a very tiny part of the overall report, making up about a quarter of shipments and orders. Unfortunately, the news here is not so great. Orders have been down for each of the last four months and even the year-over-year growth rate is slipping again.

Housing Starts Report Delayed
Just when I was hoping to get some valuable data on the market for brand-new homes, the U.S. Census Bureau pulled the rug out from underneath us. The bureau's statistician could not pull together all the data necessary for the starts report because the government furlough interfered with the data-collection process. (The housing starts report is more labor-intensive than most reports because it requires visits to a sampling of housing sites.) Therefore, the government has decided to release the data for September, October, and November all at once on Dec. 18. That means three months without results in housing starts (August data was released in mid-September). This certainly won't please the Fed, which has promised to be very data-centric regarding its decision to taper bond and mortgage purchases.

Permits Were Released, With Mixed Results
Permits, which do not involve field visits, were released, however. Permits are required before homebuilding can begin in most locations, but not every permit turns into a start, and the leads and lags between permits and starts are quite variable. Still, permits are a great leading indicator of the housing market, but it is starts that drive economic activity calculations, such as GDP. The headline permits number was stunning at 974,000 units for September and 1,034,000 units for October, the best level of the recovery and well above August's 924,000-unit level.

Permits Not Much Above Average
Although many were excited by the data, there were a few downsides. First, the permits data has been relatively volatile, especially this year, as the rapid increase in mortgage rates drove buyers to close existing homes faster. With some leveling of rates, some bounce-back was surely in the cards at some point. Although starts look good compared with recent months, they have barely budged from the year-to-date average of 949,000 units.

Second, almost all the big spikes in both September and October were related to multifamily home sales and not single-family home sales. Permits for single-family homes have been basically unchanged for several months, while multifamily sales have jumped all over the place.

In general, multifamily homes have shown a lot more recovery than single-family homes. Single families are at just one third of their pre-recession high, while multifamily homes are at a much more respectable 71% of the previous high. Multifamily homes aren't "bad" from an economic standpoint, but they are different. Prices might not vary much between the two categories, but multifamily homes are less labor-intensive, meaning fewer jobs per unit. The smaller size might also mean less furniture and other goods are purchased with condos than a standalone home.

Finally, the absolute number of permits is looking strong and at a recovery high, but the year-over-year growth rate is going in the wrong direction, suggesting at least some diminution of housing growth rates. That means that housing's contribution to growth is likely to diminish in the months ahead.

Home Prices Still Moving Up, Just More Slowly
This week, both Case-Shiller and FHFA home price indexes were released. ( CoreLogic data was released earlier in the month.) FHFA and CoreLogic both showed slowing in home price increases while the Case-Shiller data was slightly stronger, and maintained a relatively healthy rate of improvement. The FHFA data tends to be a lot less volatile in both directions and represents a more diverse cross-section of homes that are guaranteed by a government agency. With some pushback from buyers due to affordability and higher mortgage rates, a slowing in price increases is not unexpected. In fact, general expectations are for home prices to rise about 5% (based on the Case-Shiller 20-City Index) in 2014, considerably below this year's rate.

Regional Variances in Home Prices Persist
The home-price fairy is not treating every part of the country the same. Despite a nationwide price increase of 0.3%, six out of nine regional markets were either flat or declined, according to the FHFA data. Only the Pacific, East South Central, and South Atlantic regions managed to eke out an increase. For the year-over-year data, only the Pacific and Mountain regions grew by more than 10% (19.2% and 11.5%, respectively) while every other region grew at a slower but more sustainable mid-single-digit rate.

Case-Shiller data showed the same skewing. Though year-over-year prices are up just 13%, Las Vegas is up 29%, San Francisco 26%, Los Angeles 22%, and San Diego 21%.

Pending Home Sales Fall Short of the Mark
Pending home sales, an excellent predictor of existing-home sales and the overall economy, fell short of the mark again in October. Month-to-month pending home sales were down 0.6% versus expectations of an increase of more than 1%. The year-over-year data looked even worse with single-month data down about 2% year over year.

Even the three-month average year-over-year data narrowly missed an outright decline, as shown below.

Although existing-home sales have fallen too, the slowdown in existing-home sales has not been as abrupt as the fall in pending home sales, and the data suggest some bad months ahead for existing-home sales. Existing-home sales' positive impact on overall economic activity should diminish in early 2014 because of lower broker commissions, furniture sales, and remodeling activities.

National Association of Realtors Projects No Growth in Existing-Home Sales in 2014
The National Association of Realtors predicts that higher mortgage rates and higher prices (which combined mean less affordability) as well as continued tight lending policies will hold the market back in 2014. It is also projecting that the price of the median home is likely to grow at 5% versus a projected 11% rate for 2013. Slower price growth is probably a good thing, but flat home sales is not a good omen for the economy, for the reasons noted above.

Weekly Shopping Center Data Tough to Interpret This Week
On the surface, weekly shopping center data appeared to have another week of stodgy 2.1% year-over-year growth after better data last week. However, that is comparing a holiday week last year with a nonholiday week this year. They are supposed to be seasonal adjustment factors, but I am not so sure that they can capture the almost full-week shift of the Thanksgiving date (Nov. 22, 2012, and Nov. 28, 2013). For now, it looks like retail is holding up, but I continue to hear stories of substantial discounts early in the season. I predict it will be a good holiday season for consumers, but not so much for retailers.

Payrolls, Consumption, GDP, and Auto Sales All On Tap for Next Week
Next week is supposed to be a crazy week for data, although not all of the data is that important. And even the typically more important data may be plagued by data-collection issues and the up-and-down swings caused by the on-again off-again impact of the government furlough. Hurricane Sandy also affected some the data from a year ago (both September and October), distorting some year-over-year comparisons.

Consumption Report Could Shed Light on the Soft Services Sector
I would probably single out the consumption and income report as the most important. At 70% of GDP, consumption is the largest component of the GDP calculation. Some data points we have for October suggest some slightly better consumption numbers that will look even better when factoring in deflation. Goods have been doing better than services lately. Durable goods did particularly well in the third quarter. Interestingly, the services side of the economy has been a little weak lately, so I will be taking a particularly hard look at the services statistic. Durable goods did well in the third quarter but are showing some signs of slumping early in the fourth quarter. The consensus is looking for consumption growth of 0.2% month to month and income growth of 0.3%, which both seem reasonable, but certainly don't suggest a new boom.

Employment Report Could Be Just a Little Softer
The employment report is usually a tad backward-looking and generally isn't a great indicator of future activity. However, this particular report is important because it will be the last report before the Fed's next meeting in December, when it will revisit its decision to taper bond purchases. It has cited employment data in general, and the unemployment rate in particular, as key guides in its decision-making process.

Overall employment growth is expected to slow modestly in December from 204,000 jobs added to 180,000 jobs added. The more important year-over-year, averaged employment growth rate in private-sector employment is likely to remain at 2%. I have not been particularly impressed with the employment report over the last several months, as wage growth and hours worked growth have been modest at best, despite respectable overall employment growth. Key growth sectors have been restaurants and leisure, the lowest-paid sectors in the report. Better-paying finance, manufacturing, and health-care job growth have moderated or declined modestly.

November Auto Sales Data Might Not Be That Useful
Auto sales have been on a yo-yo with strong summer sales followed by a slow fall, partially due to the government shutdown and partly due to the way the Labor Day holiday sales were counted. November is not a very good month for auto sales, but December is one the biggest months of the year. As a short month with a holiday and bonus payments yet another month away, November sales are not indicative of much. Yet, the consensus estimates are relatively optimistic at 15.6 million units, which would be a nice rebound from October's mediocre 15.2 million units. Anything better than that would be indicative of a much more confident consumer.

Lots of Other Releases, Too
Though none is particularly market-moving (unless there is a massive surprise), there are a lot of other releases next week. Those include construction, ISM purchasing managers data, new-home sales, trade balance, and the second GDP revision for the second quarter (with expectations for a revision up to more than 3%, which looks overly optimistic to me).