A lot of people are asking “where’s the bottom?” in terms of both housing and the stock market. A survey of Bloomberg users found that the consensus expectation for the bottom in global stocks is six months out. Well, what the heck, I’m a Bloomberg user (though they didn’t ask me), so I’ll take a stab at it myself.
For starters, let’s take a look at the National Association of Home Builders’ (NAHB) Housing Market Index, a gauge of builder sentiment. The index peaked at 72 in June 2005, when housing starts were in excess of 2 million units, annualized. A sharp decline then ensued, arrested only by a one-month pickup in October 2005, a month of no change in January 2006, a five-month pickup from September 2006 through February 2007, and another unchanged month last November. Other than those months, the index plummeted to a low of 18 in December of last year, before rebounding to 19 last month.
These numbers are record lows for the index, which is based on three components: current sales, projected sales six months hence, and prospective buyer traffic. The current lows are lower even than during the last credit-driven recession, in 1991.
Now, the recent numbers suggest that builder sentiment may be bottoming out. It fell from a September 2007 reading of 20 – matching the then-record low from January 1991 – to 19 the following month, where it stuck before dropping to the new record low of 18 in December, then rebounding to 19 last month. The forecast for February’s reading – due out next Tuesday – is also 19. So in other words, the index has been hovering around the 18-20 range for six months, if the forecast holds.
So, has it bottomed out? Well, one could make that argument, especially if the forecast does come to fruition. I won’t make that claim, though, for a couple of reasons. First, the forecast could prove overly optimistic, and I’d look stupid for taking a position three days before a number comes out. I hate when that happens. And second, many observers thought the bottom had been reached in September 2006, when the index hit 30, then rebounded to 39 by February, before resuming its free-fall. Boy, were they wrong. (For the record, I kept stalwartly predicting that the worst was yet to come, and turned out to be right, looking pretty smart in the process. I love when that happens.)
Part of the index’s recent performance has been driven by an uptick in the six months’ forward sales outlook. So apparently builders think that sales will improve beginning in about May. If they’re right, we could say that housing will officially bottom out this spring or summer.
That may be premature. Let’s look at how starts and permits have behaved. Starts peaked at about 2.3 million units, annualized, in January 2006, a couple of quarters after the peak in the NAHB Index. Thus began a slide that has seen starts plummet to just more than 1 million units, less than half the peak (that number, too, is due for release next week, and may well see a piercing of the 1 million unit threshold, which hasn’t happened since 1991).
Similarly, permits peaked at almost 2.3 million in September 2005 (leading starts, as would be expected) before beginning a similar decline that has also brought us to the brink of the 1 million permit threshold. If anything, the recent performance of both starts and permits has exhibited an acceleration of the decline. And the continued slide in permit issuance suggests that starts will also continue to fall.
It should be noted that in 2002, the National Association of Realtors, using readily-available and reliably-forecasted demographic and immigration data, projected that starts would need to average 1.5 million units per year throughout this decade to keep up with housing demand. Well, starts have never been at or below that level since the projection was made, averaging about 1.8 million from 2001-04, and never falling below 2.0 million in 2005.
As a result, the supply of new homes for sale has risen from 4.1 months’ worth in mid-2005, when the NAHB Index peaked, to 9.6 months in December of last year. And it keeps going higher, in spite of the fact that builders have dramatically cut back on building.
But given the excess of construction activity over projected demand as noted above, by my estimation we built about 1.9 million too many homes from 2001 through the end of last year. To work off that inventory, builders could just sit completely idle for the next five quarters. Or, at a level of, say 1 million starts a year, it would take about four years to work down the inventory. Either way, it’s hard to argue that we’ve seen the bottom in housing yet. The reality is probably something in the middle – in other words, between five quarters and four years, with starts between zero and 1 million. So a reasonably safe guess for the bottom might be two and a half years, with starts falling to about 750,000 a year or so at the trough.
One indication that things could hit bottom somewhat sooner is the modest improvement in the supply of existing homes, which rose from 4.5 months in 2005 to 10.7 months last October, before contracting to “just” 9.6 months in December. But at that pace it would still take more than a year to get the supply of existing homes back down to historical norms. Then, we could see trade-up activity start to take a serious bite out of the excess inventory of new homes. That would still probably put the bottom two years out.
Working against this, of course, is the continued foreclosures of subprime ARMS, and now serious delinquencies and foreclosures on prime mortgages are trending up as well. Also, this year will bring “option ARMageddon,” as I argued earlier, which will only result in more vacant homes on the market. And there is nothing the Fed can do about this, short of cutting the funds target back to 1.00% by spring, in which case I’m leaving the country.
So where’s the bottom for housing? None of this is definitive, but I’m pretty confident it’s not this summer. I’m not even sure it’s foreseeable, as there’s another, longer-term, systemic housing correction coming in 2010, one related to the demographic shift among the baby boom generation. But that’s a topic for another day. For now, I’ll say the bottom of this correction in housing will occur around the end of 2009, at best.
As for the stock market, let’s look at the S&P during and following the last recession. It peaked above 1,500 in the second half of 2000 before beginning a slide that persisted through the subsequent recession, which lasted only a couple of quarters of the following year, but it extended beyond that through 2002, bottoming out at just over 800 that October. The 9/11 terror attacks didn’t help, but they didn’t artificially depress stocks for long either; the S&P did plunge right after 9/11, but rebounded through the end of 2001 before resuming its slide early the next year.
So, the duration of the S&P’s slide in the last bear market was about two years, during which time it fell by almost half, from above 1,500 to 800.
More recently, the S&P peaked – again above 1,500 – last October, and has since shed just over 200 points, or about 13%. We could just assume the slide will extend another 20 months or so, and about another 500 points, like last time. But we’d like to have something a little more concrete than that.
Well, here it is. First, this recession will be worse than the last one. It will last longer, and run deeper, especially as housing’s woes extend. So the concomitant decline in equities will likely last longer as well, and run deeper. That’s just the seat-of-the-pants assessment, though.
A second observation can be made by returning to the NAHB Index. As the 1990-91 recession loomed, George Bush Sr. commented, “As housing goes, so goes the economy.” A pretty astute observation. The NAHB Index tracks pretty well with the S&P 500, lagged about 18 months (for more detail on this, go to www.cnbsnet.com, click on “View all MarketCast Articles,” then click “Search MarketCast Articles” at the bottom of the page, then select May 2007 from the “View a Specific Issue” drop-downs, and click on the article titled the same as President Bush’s quote above).
If we look at what’s happened to the NAHB Index subsequent to that article, which was penned during the early 2007 rebound in the index (and I did a follow-up in the December MarketCast titled “Recession Indicators Revisited”), we can extrapolate the expected trailing S&P trend. If we assume that the NAHB Index has indeed hit bottom recently, then the bottom for the S&P would be approximately a year and a half out. That would correspond fairly well with the rough 20-month estimate postulated above.
As for the trough level, that same exercise in extrapolation puts the S&P at about 400, or about 1,000 points lower than today. That falls somewhere between the 800 we get from just looking at the last bear market, and the seat-of-the-pants assessment that things are worse this time. One thing’s for certain: stocks won’t meaningfully rebound until the housing crisis ends, and that’s about two years out.
So is 400 my forecast? I’ll hedge my bet, and say the bottom’s around 600, and it’s a good two years out – call it second quarter 2010. Now, aren’t you glad you read this far to reach such a gloomy conclusion? Told you I was a curmudgeon.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment