This week was all about the government shutdown and the even more worrisome prospect of violating the debt ceiling sometime later this month. A debt ceiling violation could lead to the deadly combination of higher interest rates and a slower economy. In a world where markets swoon on a 20,000-job miss in monthly employment data (on monthly job growth of about 200,000), losing 800,000 government jobs in one shot is a very big deal.
Although the market has been soft in recent weeks, it seems to have hardly grasped the potential of a longer-term shutdown. We have reached this crisis point so many times recently with no negative consequences that the market seems almost numbed to the potential pain. And I have to admit, if they come to a settlement on the debt ceiling and budget issues in the next week or two (and include retroactive pay for government employees), the economy will do just fine. But even a month-long shutdown would likely cut GDP growth by at least 0.5% in a world of 2% growth. In our weekly video, we discussed the impact of the shutdown.
The other negative consequence of the shutdown: The government agencies have stopped releasing statistics. This week we missed the all-important employment reports and construction data for August. (Without much data, I'm not sure what I'll write about next week.) More seriously, I am not so sure how the Federal Reserve will make its monetary decisions without the hard data it said it was craving just a few weeks ago.
The other economic news this week was consistent with a slowing, or at least plateauing, economy. ADP job growth missed the mark, heretofore strong auto sales went soft, and home price growth plateaued. Also, anecdotal evidence suggests that layoffs might be on the rise. Recently we have heard of layoffs at BlackBerry BBRY, mortgage processors, fixed-income trading firms, and a big layoff at Merck(MRK). Health-care layoffs are up, too, according to Challenger Gray's most recent survey.
On the positive side of the ledger, weekly retail sales growth managed a 2% increase after falling below 2% the previous week. Furthermore, despite the recent headlines, initial unemployment claims remained very near recovery lows. Meanwhile, retailers, with the exception of Target (TGT), seem to be ramping up holiday hiring intentions sharply from the same period a year ago, even in the face of a soft holiday shopping forecast.
Auto Sales Cool, but Nearly Impossible to Interpret
Autos slipped from August's stellar 16.07 million seasonally adjusted annual rate to 15.3 million, according to Automotive News (official government data is not available because of the shutdown). However, the seasonally adjusted days sales adjusted SAAR rate of 15.3 million units is still higher than last year's 14.8 million units. Without the seasonal and days sales adjustments, auto sales were down 4% year over year.
The early Labor Day holiday pushed holiday weekend sales into the August data, artificially inflating that report and depressing the September report. An average of the two months, August and September, is probably a better indicator of the state of the auto industry, and that figure is about 15.7 million units. Nevertheless, combining August and September suggests growth of about 7% annually, down from the 10%-11% growth rate of earlier months this year. Unfortunately, the trend was looking a little worse at the end of September, when sales near major government installations began to peter out, according to dealers.
A lot of consumer and retail-oriented data has looked weak since July, while the auto data appeared to buck the trend. Now the auto data seems to be confirming at least a little slowing.
Labor Data Mixed This Week; Employment Report Missing in Action
Unfortunately, the Bureau of Labor Statistics' official employment report was one of the casualties of the government shutdown. However, Challenger Gray layoff data, low initial unemployment claims, and improved retail hiring plans all pointed to a steady and perhaps somewhat better employment situation. Still, the ADP version of the employment report showed more of the slow, unsatisfying growth rates we have seen for the last several months. If all this data is indicative of a mediocre employment report, I don't think the Fed would be compelled to end its bond-buying programs quickly.
I discuss the ADP report in more depth in one of my videos this week. That report showed little change in recent trends, with private sector jobs growing at 166,000 jobs, up from 159,000 jobs added in August (ADP data). The report was a bit of a disappointment, as the consensus estimate was for 180,000 jobs to be added. The ADP report has been a less-than-perfect indicator of the official government report in the short run, but at least gives us some hope that things weren't falling apart.
As discussed below, the manufacturing sector grew a tiny 1,000 jobs while the trade sector grew an impressive 54,000 jobs, indicating continued increases in the retail sector despite lackluster sales. Construction was another bright spot with 16,000 jobs added. However, the financial sector lost 4,000 jobs as the layoffs of mortgage processor and trading personnel began to kick in. Health-care workers were not reported separately by ADP, but Challenger Gray layoff data suggests that health care was also soft.
Home Price Appreciation Plateaus, Monthly Data Show Seasonal Slowing
CoreLogic (CLGX) kicked off the home price data releases for the month of August with single-point growth rates of 12.4% year over year and 0.9% July to August. This marks the 18th month in a row of home appreciation. Since February 2012 (the recession low for home prices) the CoreLogic index has increased by a remarkable 23%. However, the CoreLogic index is still 17% below its peak in April 2006. Speaking of peaks, a total of 11 states, mostly in the central part of the country, are now at new all-time highs. While many of these are small states, Texas, Colorado, and Washington, D.C., made the list of new highs. Two additional states are within 1% of their all-time highs.
With increases in family earnings and lower interest rates, I am not that worried about approaching old peaks, at least not yet. Sharp price appreciation remains concentrated along the West Coast. However, this month the top five states for home price appreciation included Georgia and Wyoming. Just five states, all non-populous ones, had growth below 2.75%. No state showed a year-over-year decline.
I do like looking at a three-month average of the annual data and to peek at the monthly increases, which I show below.
There is clearly some slowing either because of seasonality, buyer fatigue, or higher interest rates. Home price growth accelerated from about 4% last August to 11% by May and really hasn't budged much since. The slowing is more pronounced in the monthly data, but some of that is definitely related to seasonality.
Housing Affordability Slipping Again, Down 27% From Best Levels
The National Association of Realtors compiles an affordability index each month that includes mortgage rates, median home prices, and median family income as inputs. Lately, I am not a huge fan of the index because it uses median home prices, which are strongly influenced by the mix of homes being sold. When cheap bank foreclosures dominated the index, affordability looked better than it really was. Now that those really cheap homes are off the market, homes don't look nearly as affordable. The median home price in the Realtor survey is up 37%, whereas the mix-adjusted CoreLogic data shows home prices up just 23% from cycle lows.
With that warning, I note that affordability has still slipped a lot. Affordability in August was down 27% from its best level, which was reached in January 2012, when home prices were near their lows. Surprisingly, all of that deterioration comes from increased home prices and the above-mentioned mix issues, with rates in January 2012 equal to today's 4.4%. I was modestly surprised in looking at the data that mortgage rates were only below 4% for just over a year, from May 2012 to June 2013, according to the Realtors Association. Likewise, incomes moved up only a little and improved the affordability index by just a couple of percentage points. So while everyone is whining about higher interest rates, it appears that home price increases, and not interest rates, are really impinging upon home sales.
September Pending Home Price Data Shows Even More Slowing; No Need to Panic
Looking ahead, CoreLogic also projects September data based on pending home sales in its database. Final results based on closed sales have generally been coming in just a smidge higher than this early warning index. In any case, the year-over-year single-point estimate for September was a still-healthy 12.7%; however, the month-to-month growth rate slowed to just 0.2%. With some massive increases in prices in a short period of time combined with higher mortgage rates, some slowing in price appreciation is to be expected. As long as mortgage rates don't shoot drastically higher or the economy slips badly, I don't think real estate prices will register any meaningful declines.
ISM Purchasing Manger Data Grows Again, but I'm Not Sure It Means Much
Purchasing manager data has proven to be a real market mover ever since it was one of the very few metrics to successfully project the start of the recovery back in 2009. The regional PMI data can also prove to be market-moving. However, the popularity of the national index comes at a time when its predictive ability has fallen off a cliff. The statistical fit between the industrial production statistics for manufacturing and the composite ISM index has been excellent both for the last 15 years and since 1949. However, since 2011, there has been next to no predictive value to the ISM data. None. Thanks to my colleague Adam Fleck of Morningstar's industrials team for running those correlation statistics for me. He volunteered after I looked at the tabular data below, which, on the surface, appeared to show no relationship between industrial production and the ISM index.
However, what really got me thinking about the ISM data was ADP employment data for September, which produced a measly 1,000 jobs added in the manufacturing sector, despite the alleged boom in manufacturing. So unfortunately, it looks like we have to throw the ISM index into the worthless data bucket that includes the Baltic Dry Index and the University of Michigan Consumer Sentiment Index. At least for now.
No Concrete Reason for the Shrinking Relevance of the ISM Data
However, for quick reference I will note that the ISM Index is now at 56.2 for September, its best reading since April 2011. The new order index looks even better with a 60.5 reading, which suggests even more good news ahead. I suppose that the decent ISM readings of late are at least indicative of manufacturing not falling apart. Why the reading has gone haywire recently is a good question that I can't answer. I suspect shifting auto sector seasonality might be part of the explanation. I also suspect that there are fewer respondents that actually take the time (or have the time) to calculate numbers to determine if things are better or worse, instead relying more on a gut feel and anecdotal evidence. I have also worried forever that the auto industry seems to have an unusually large impact on the data, despite the industry's relatively small size. More just-in-time inventory systems might be messing with the data, too.
Suspending the Week Ahead Section for Now
With the government sharply curtailing data releases for now, I doubt that many reports will see the light of day next week. With the exception of the retail sales report, none of next week's "normal" reports would be classified as market moving. Non-government releases due include the NFIB Small Business Confidence Report and the University of Michigan Consumer Confidence survey--not exactly big market movers.
- Budget, Tax & Economy
- interest rates
- government shutdown