CMI: Absolute Demand and Disposable Income Decline

by Rick Davis

We have recently called our readers’ attention to our Daily Absolute Demand Index, which compares the current on-line consumer demand for discretionary durable goods over the past 60 days to the same consumer demand in 2005:


The baseline level of 100 represents the average consumer demand for the 2005 calendar year, and our data has been normalized as best we can for the inexorable movement of market share from brick-and-mortar to the internet (using our proprietary “same shopper” methodologies). The plunge in the above chart started on October 27th, and to-date it represents a drop of over 10% (in the admittedly volatile demand for discretionary durable goods) from the recent peak level recorded on October 26, 2011. It also marks a return to levels more typical of the period from from May 2010 through June of this year.

It is important to remember that the base-line number of 100 represents the year 2005, more than a year before U.S. private sector real per-capita wages peaked during the first quarter of 2007 (per line 4 in the BEA’s Personal Income and Its Disposition, Table 2.1, converted from nominal data to “chained 2005” dollars and then to a per-capita basis). By the first quarter of 2010 those same per-capita real private sector wages had contracted by an aggregate -11.2% — and even as late as the BEA’s latest report they remained -8.6% below that peak, now some four and a half years ago.

The rough timing and consequences of that real private sector wage contraction can be seen in our Monthly Absolute Demand Index, which captures the leveraged impact of real wage losses in the volatile and marginal spending of U.S. consumers:


The good news in the above chart was the partial recovery of demand during the late second quarter and the majority of the third quarter of 2011. The bad news is that the BEA has subsequently reported that during 3Q-2011 real per-capita disposable income for all Americans again contracted by a -1.7% quarter-to-quarter rate — or a -6.6% annualized rate if you prefer more brutal headlines. That quarter-to-quarter contraction in
disposable income is likely the driving force behind the recent drop in consumer demand that we have so clearly documented in the top chart, but which will probably show up in other data sources and the main stream media only weeks or months from now.

Our synopsis reading is that whatever comfort the consumer was feeling from dropping gasoline prices late in the second quarter has largely dissipated under the continuing pressure of contracting disposable incomes. The problem with synopsis readings is that the economic environment will remain complex, as each household reacts to contracting disposable income in their own unique fashion — causing the macro-level impacts on the various segments of the consumer economy to behave in unpredictable ways.

The next few quarters will indeed be interesting times, bringing the ancient Chinese curse freshly to mind.

The Usual Suspects

Meanwhile, our year-over-year Weighted Composite Index has moderated somewhat from the all-time record high that we recorded August 13, 2011 and at least plateaued over the past 11 weeks:


That moderation has been greatly assisted by a corresponding fall-off of our heavily-weighted Housing Index once again into year-over-year contraction, as seen in this day-by-day view:


That fall-off in the housing index has been caused by the flattening out of the weekly data for new loans for newly purchased residential units:


— and the demand for refinancing of existing mortgages:


Both of these “topping out” actions seem to indicate that whatever pop may have been provided to the housing industry as a result of historically low mortgage rates has waned, perhaps as a result of the Fed’s promise to maintain the low rates — thereby removing any incentives to move purchases forward in the face of the continued softening of housing prices.

At the same time the mortgage-rate-independent demand for apartments has also clearly maxed out:


This might also reflect a temporary lull in the demand for rental housing as the duration of foreclosure proceedings continues to grow.

And the Automotive Index remains up a healthy 6% year-over-year:


We have mentioned before that to some extent autos are neither discretionary nor durable over the long haul, meaning that they ultimately have to be replaced at some point in time. This non-discretionary component of auto demand, coupled with the probable use of auto loans as a tool for rehabilitating damaged household credit, has provided good news for a select subset of the domestic car makers:


Note that the majority of the good news in that chart has come from the Ford and Chrysler badges, which have more than offset the anemic growth shown by General Motors.  And we continue to suspect that demographics is driving the continued year-over-year growth of the Korean badges, as the younger car buyers without long-established brand loyalties flock to the perceived value of the Korean brands and pull market share with them:


All that said, we return to one of our principle themes:  this “Great Recession” has not been an experience that has been fully shared among all portions of the American public.  To that end we offer the recent movement in the Luxury Autos chart:


The Hazards of Year-Over-Year Charts

Any of our year-over-year charts are simply that: they show the changes in on-line consumer demand for discretionary durable goods relative to the same commerce date one year ago. Historically, information about day-to-day changes in the year-over-year behavior of consumers in the particularly volatile discretionary spending arena was useful because it provided an amplified signal when meaningful trends were just starting to develop.

Needless to say, that same amplification can play havoc with intuition when the economy is so volatile that amplification is no longer required. For example, should the relative consumer demand for widgets decline from 100 to 50 during one year, only to return to 100 the next, year-over-year measurements would initially show a 50% contraction (that makes sense) during the first year, followed by a 100% growth the next (which seems at first blush to be outrageously high, given that everybody knows that total demand has only returned to just “normal”). But the math is simple: a 50% loss followed by a 100% gain only puts you back to where you began.

Secondly, year-over-year numbers also put you at the mercy of extraordinary events in the prior year.  In the current case of our Weighted Composite Index the numbers look good
because they are being compared against horrific numbers from last year — as will be the case through the second quarter of 2012.

For all of these reasons we are busy building a completely new set of “Absolute Demand” indexes that at the end of the year will supplant the current set of year-over-year charts. All will share the same 2005 baseline year used by our current aggregate Absolute Demand indexes. And at the same time we will be bringing a number of new data sources on-stream that will provide even more robust sampling.

This is the first major change in our methodologies in the past several years, and we are excited to be able to improve both the intuitive and statistical quality of our data at the same time. We will be keeping you informed as the transition period approaches …

Related Articles

All articles about Consumer Metrics Institute