by Calculated Risk on 5/24/2013 01:13:00 PM

By request, I've updated the graphs in this post with the most recent data. Last year when I wrote The Housing Bottom is Here and Housing: The Two Bottoms, I pointed out there are usually two bottoms for housing: the first for new home sales, housing starts and residential investment, and the second bottom is for house prices.

For the bottom in activity, I presented a graph of Single family housing starts, New Home Sales, and Residential Investment (RI) as a percent of GDP.

When I posted that graph, the bottom wasn't obvious to everyone. Now it is, and here is another update to that graph.

Click on graph for larger image.

The arrows point to some of the earlier peaks and troughs for these three measures.

The purpose of this graph is to show that these three indicators generally reach peaks and troughs together. Note that Residential Investment is quarterly and single-family starts and new home sales are monthly.

For the recent housing bust, the bottom was spread over a few years from 2009 into 2011. This was a long flat bottom - something a number of us predicted given the overhang of existing vacant housing units.

We could use any of these three measures to determine the first bottom, and then use the other two to confirm the bottom. These measure are very important and are probably the best leading indicators for the economy. But this says nothing about house prices.

The second graph compares RI as a percent of GDP with the real (adjusted for inflation) CoreLogic house price index through February.

Although the CoreLogic data only goes back to 1976, look at what happened following the early '90s housing bust. RI as a percent of GDP bottomed in Q1 1991, but real house prices didn't bottom until Q4 1996 (real prices were mostly flat for several years). Something similar happened in the early 1980s - first activity bottomed, and then real prices - although the two bottoms were closer in the '80s.

Now it appears activity bottomed in 2009 through 2011 (depending on the measure) and real house prices bottomed in early 2012.

by Calculated Risk on 5/24/2013 10:29:00 AM

Last week I summarized some recent research on the labor force participation rate. The following piece from Michelle Meyer at Merrill Lynch argues the LFPR will likely move sideways over the next few years. Changes in the participation rate have important implications for the number of jobs needed to lower the unemployment rate.

An excerpt from Michelle Meyer's piece:

The future trajectory of the labor force participation rate (LFPR) is very important in gauging the trend in the unemployment rate and risks to wage inflation. In order to forecast the labor force participation rate, we must understand the drivers behind its recent sharp movements – to what extent is a long-term trend related to demographics (aging population) versus secular or cyclical dynamics?

We can isolate the effect of demographics on the LFPR by looking at the participation rates by age cohort. The aggregate LFPR is equal to the summation of each individual age cohort's LFPR weighted by its share of the population. ... This suggests that half of the 2.7pp decline in the LFPR since the onset of the recession can be explained simply from the aging population. In other words, holding all else equal – meaning no business cycle dynamics – the LFPR would be at 64.6% today compared to the actual rate of 63.3%. The remaining 1.4pp drop is due to some combination of secular and short-term cyclical factors.

The two primary secular trends are the decline in the LFPR among the youth population and the rise among 55+. The LFPR for 16 to 19 year olds plunged to 34.3% last year from 52% in 2000. While this may have been accelerated by the past two recessions, we believe this is a permanent trend. On the other end, the LFPR of the older population has increased, likely reflecting higher life expectancy, less confidence in social benefit programs and loss of wealth from the Great Recession.

... we still believe that there are some cyclical components. One way to gauge the cyclicality of the LFPR is to observe state-level variation in the relationship between the LFPR and the health of the economy. Based on work from a recent San Francisco Fed paper, we compare the percentage decline in state payrolls to the decline in the LFPR during the recession, both weighted by the relative size of its labor force. We find a positive relationship where larger declines in employment are associated with bigger drops in the LFPR. The paper does the same exercise for prior recessions and finds a positive relationship existed in each, with the exception of the 2000 cycle.

If the relationship holds on the downside, do we also observe it during the recovery? There is little evidence of such correlation in this recovery, but it does exist for prior cycles. In prior cycles, the positive correlation did not become apparent until the economy had exceeded the previous employment peak by a significant amount. This suggests that the cyclical pressure in the LFPR may not be observed until 2015, at the earliest.

We can simulate a future path for the LFPR based on our assessment of the drivers of the downturn in the LFPR. The first step is to account for the continued aging of the population using the Census Bureau's projections by age cohort. If we keep the LFPR by age group constant at 2007 levels and only allow for demographic adjustments, we find that the LFPR will fall by another 3.3pp by 2025 and then slip to 59.2% in 2050. ...

We also assume that the secular trends exhibited during the last decade persist, but at a slower pace, implying a modest downturn in the LFPR among the youth and an upward trajectory for the 55+ age group. ... Based on these rough assumptions, we forecast the LFPR will slip slightly this year, but with a stronger recovery under way next year, the LFPR should start to level off some and potentially increase beginning in 2015 (Chart 4).

The cyclical dynamics, in our view, are not strong enough to generate a pop higher in the LFPR given the downward pull from demographics. But at a minimum, we expect these dynamics can counter the downside pressure and allow the LFPR to move sideways once the recovery builds momentum.

by Calculated Risk on 5/24/2013 09:06:00 AM

From the Department of Commerce: Advance Report on Durable Goods Manufacturers’ Shipments, Inventories and Orders April 2013

New orders for manufactured durable goods in April increased $7.2 billion or 3.3 percent to $222.6 billion, the U.S. Census Bureau announced today. This increase, up two of the last three months, followed a 5.9 percent March decrease. Excluding transportation, new orders increased 1.3 percent. Excluding defense, new orders increased 2.1 percent.

This was above expectations of a 1.1% increase. This report is difficult to predict and very noisy month-to-month.

by Calculated Risk on 5/23/2013 08:32:00 PM

Several people have asked me about this article at CNBC by Jeff Cox: Why Housing Won't Drive the Recovery

Despite data points that in some cases are at multiyear highs, Robert Shiller, Karl Case and David Blitzer believe there are multiple headwinds that will keep a lid on housing gains.

Among the obstacles are a low level of new housing starts, an unexpectedly slow migration of so-called shadow inventory onto the market, and continued difficulty for buyers to secure financing.

"You've got a lot of breathless commentary in the media," said Shiller, a Yale University economist. "All this talk that we're in this great recovery—we probably are in the short run, the longer run doesn't look so terrific to me."

First, housing (technically residential investment) will be a key driver for the economy. Period. 

It is important to understand that "residential investment" is mostly new homes and home improvement. For existing home sales, only the broker's commission is included in residential investment (nothing is added to the housing stock).  Those looking at the level of existing home sales are looking at the wrong number, as are those focused only on house prices.

Look at "headwinds" that are mentioned in the article:
1) "a low level of new housing starts".  That is a headwind?  To me, the low level of starts means there is more upside based on demographics.  The homeownership rate peaks for those in the 55 to 75 age group, so the boomers will not negatively impact homeownership for a decade or more.

2) "an unexpectedly slow migration of so-called shadow inventory onto the market".  Do they expect the pace of foreclosures to increase?  I don't.  The process is very long in most judicial states, and I don't expect another wave of foreclosures hitting the market - but I do think we will see distressed sales for years.

3) "continued difficulty for buyers to secure financing".  OK, but this has been a headwind for the last couple of years.  Looking forward, I expect some loosening in lending standards.  So this is really a potential tailwind.

Nothing in this article changes my view.

Friday economic release:
• At 8:30 AM ET, Durable Goods Orders for April from the Census Bureau. The consensus is for a 1.1% increase in durable goods orders.

Note: The bond market will close early Friday at 2PM ET. The stock market will close at the normal time. All markets are closed on Monday in observance of Memorial Day.

The Calculated Risk blog is always open!

by Calculated Risk on 5/23/2013 03:33:00 PM

From Freddie Mac today: Mortgage Rates Continue Upward Trend

Freddie Mac (OTCQB: FMCC) today released the results of its Primary Mortgage Market Survey(R) (PMMS®), showing fixed mortgage rates trending higher for the third consecutive week and putting pressure on refinance momentum. ...

30-year fixed-rate mortgage (FRM) averaged 3.59 percent with an average 0.7 point for the week ending May 23, 2013, up from last week when it averaged 3.51 percent. Last year at this time, the 30-year FRM averaged 3.78 percent.

15-year FRM this week averaged 2.77 percent with an average 0.7 point, up from last week when it averaged 2.69 percent. A year ago at this time, the 15-year FRM averaged 3.04 percent.

Click on graph for larger image.

This graph shows the the 30 year and 15 year fixed rate mortgage interest rates from the Freddie Mac Primary Mortgage Market Survey®.   Not much of an increase recently ... but it will slow refinance activity.

The Freddie Mac survey started in 1971 and 30 year mortgage rates are still near the record low set last November.