by Brian Pretti CFA, Contrary Investor. September 28, 2007
We all know far too well by now that late last year and early this year, many a Fed and Treasury official were proclaiming from on high that sub-prime mortgage credit problems were contained. The party line was that problems in that particular credit sector neck of the woods were not about to spread or cause further problems in any other part of the domestic, let alone, global credit markets. Right. Unfortunately for far too many institutional credit market investors as of late, wrong. I've been documenting and discussing this lack of containment issue for many a moon, as well as factually documenting the fact that there is simply no way the actual housing market is anywhere near a bottom, although it’s certainly continuing down that path as we speak. Time to move on to another very important conceptual containment point of the moment; a new containment issue that we believe will be very important for real world domestic economic outcomes ahead.
Let's start with a quick look at some longer-term housing data now updated through the second quarter GDP report. Below we're looking at residential fixed investment as a percentage of total GDP. The most recent had been the longest up cycle for residential investment on record. Hard to imagine it would be all reconciled in a few quarters. And so far, it hasn't. The down cycle has been playing out fast, exactly as had been the case in prior cycles. It's certainly my belief that there is plenty more to come in terms of southern exposure. At best, this measure of housing investment relative to GDP bottoms somewhere near 3.5-4% of GDP. But given the extremes to the upside in the prior cycle, my personal bet is something nearer 3%, or perhaps just a touch lower. We'll just have to see how it all plays out.
But what is important, and hopefully also to you, at the current time is that there sure as heck seems to be a growing chorus now singing yet another containment tune whose lyrics extol the message that the housing industry specifically is just not that big a part of the total US economy. Please remember that the data above is only capturing the economic value of new construction in any one period, relative to total GDP. It says nothing about the direction of prices, inventory of homes for sale, etc. So yes, new construction of residential real estate in any one period is not an end of the world number that alone will determine the fate of the entire complexion of US GDP. But this is exactly the data that many are pointing to and now suggesting that the influence of the housing sector on the total economy is contained. "It's relatively small. There is a much bigger world out there in the US economy than housing. It only accounts for currently a little less 5% of total GDP. How in the world could housing possibly throw the entire US economy into a potential recession?" You know the tune, don't you? As you'll see in the chart above, overlaid is the year over year data for rate of change in nominal GDP. Directional correlation here demands acknowledgment.
As you might imagine, this new and quite convenient "containment" theory of the moment is about as shortsighted as anything I've witnessed in a good while; about as shortsighted as suggesting that mortgage credit problems would be contained to sub prime credits only. The fact is that the influence of housing in its entirety is incredibly meaningful to the totality of the US economy, at least that's the message of historical experience. It's not just about new construction, as you know. It's about leveraging the asset, it’s about job creation in finance, sales, construction, etc. It's about retail demand in home improvement, remodel, etc. I don't need to go on and on, right? I didn't think so.
The message of the interrelationship between housing and the broad economy is really contained (no pun intended) in the following four charts that cover one heck of a lot of US GDP ground. In fact the bulk of US GDP – consumption, manufacturing, employment and consumer confidence. Influence these areas and you've taken a broad brush to the entire domestic complexion of US GDP. So as you review all of the four charts below, please look for and remember one meaningful item – in each case, housing leads. Yes, in every case. I'm using the NAHB (National Association of Home Builders Index) as a read on the character of housing, per se. Set against this are payroll employment numbers, real personal consumption expenditures, industrial production and consumer confidence. Broad enough for you? Again, in EACH case, it’s clear – housing leads. Let's start with payroll employment. Here you go.
I won't belabor the point as a number of weeks back I published a discussion documenting the leading indicators of payroll employment (not including this one) pointing downward. Turns out that was a week before the "surprising" decline in August payrolls (that should not have been surprising at all). The above chart just throws yet another log on an already open fire. The directional lead and lag influence of housing on the direction of payroll employment is self-obvious.
Next at bat is simply an update of a chart I've used in the past, just more dramatic in its current message than has been the case for some time now. And so housing doesn't affect consumer spending (PCE – personal consumption expenditures)? Better think again. The correlation here is so high, even we have a hard time believing it’s this significant. It's just a good thing that factual information leaves hope and personal opinion in a ditch by the side of the road.
Here's one I have not shown you before, but it’s high time right now. Housing has no influence on the manufacturing side of the US economy, right? Wrong. It's absolutely clear in this historical retrospective that peaks in the NAHB survey have led the year over year directional change in US industrial production. Same deal at cycle troughs – housing leads. Either the prior three housing and industrial production cycles were complete flukes, or housing indeed impacts the manufacturing side of the US economy. (Hint: It's the latter, trust me.)
Finally, the relationship between housing and consumer confidence. Since this one is a bit of an intuitive lay up, I've left it for last. C'mon, how could housing not have an influence on the consumer psyche, especially given the very simple fact that housing is the largest household asset? Maybe the correct question should be, how could it really be any other way?
The last time the NAHB survey was this low, consumer confidence was close to half the level we see today. Any guesses as to which direction confidence will be heading in the quarters to come?
So in quick fashion, there you have it. Personally, I've been hearing the "housing's influence on the US economy is contained" investment rationalization far too frequently as of late. You've probably been hearing the same. I did not believe sub prime issues were contained when this little theory was being held up as a reason for complacency, nor do I believe the reality and influence of circumstances in the housing sector are contained relative to the direction of the greater US economy looking ahead, quite the opposite. Again, point blank – housing leads. Please don't forget just how important this is and the lessons history has to teach us in the virtually incontrovertible data above. The next time you hear the "housing is contained" argument, just remember the correct response – Right.
The Low Down
As I'm sure you noted in the charts above, the NAHB survey as of the latest reading is sitting at record lows for its two-plus decade history. We must be near a very meaningful low for housing, no? That's right, no. I'm going to leave you with one last chart that may indeed be one of THE most important data relationships of the moment. One of the reasons I'm so convinced that there is much more to go on the downside for housing, and why I'm convinced no one should be underestimating the impact of housing on the broader US economy of the moment, is price. Or more correctly, lack of meaningful price reconciliation in residential real estate up to this point that I believe is surely still to come. Below we're looking at the long-term relationship between the median family home price and median family income. Pretty darn simple stuff here. Level of housing prices to income. Can it get any more basic than that?
Although this may sound like simple thinking, with all of the hoopla, sound, fury and consternation over trying to "protect" home-debtors against potential adverse mortgage credit issues, we believe the focus is completely incorrect. As you know, both Bernanke and Paulson have been lobbying to allow Fannie and Freddie to expand their balance sheets (lending), as well as raising conventional mortgage loan limits. All of the proposed short term band-aids or potential cures for mortgage credit problems de jour revolve around expansion in lending. Of course, this has been the very problem horse that has brought us to our current circumstances. As we look at the chart above, the message seems as clear as a bell. The problem is that home prices still remain too high relative to median household income levels. Of course the solution, if you will, as per this diagnosis is to allow the housing cycle to play out and existing home prices to decline to much more reasonable levels relative to family income. After all, how can the problem for housing at the moment (prior period excessive mortgage credit issuance) also be the solution (forward excessive mortgage credit issuance)? It can't.
The data used to construct the above chart tells us either one of two things plays out dead ahead. Either housing prices fall relatively meaningfully from here, or US domestic wages rise relatively meaningfully from here to get this relationship closer to being in line with historical experience. Which do you think will be the outcome ahead? If mortgage credit affordability is an issue, can it really be that housing prices are not the issue? Of course not. Although consumers have done a pretty good job hanging in there, so to speak, up until now with housing prices softening over the prior year and one half plus, the major test really lies ahead. At least since 1970, every single housing cycle saw the median housing price to income ratio fall back to what we've calculated as the average for the entire period shown. So this one will be different? I beg to differ. If I had to guess, I'd say a trip in this ratio to the 350-375% level is an extremely reasonable expectation before the current cycle has concluded, but we need to be prepared for reconciliation to go a whole lot lower. That's another 10-15% decline in median family home prices from here, at best. Can I suggest 2008 could be quite the interesting year for housing prices in the US? Can I also suggest 2008 could be quite the interesting year for the broader US economy? No wonder Bernanke chose to throw a 50 basis point rate cut ball as the first pitch of the monetary inflation world series.
© 2007 Brian Pretti