CMI Weighted Composite Continues Plunge

by Rick Davis

The year-over-year data in our Weighted Composite Index has continued a spectacular deterioration over the past month:

Click on graph for larger image.

A rational person might ask:

(1) Is that plunge real?

(2) Why haven’t we seen similar drops in retail surveys?

(3) And if your numbers are real, what on earth is going on?

Let’s answer those questions in turn:

(1) Our Weighted Composite Index numbers certainly measure real consumer behavior, but it must be remembered that the index shown above reflects year-over-year changes. Because of that the index is impacted by extraordinary changes in consumer behavior during the same time frame of the prior year. In this case, during the second half of last year the US consumer was benefiting significantly (in both pocket change and budgeting psyche) from sharply falling gasoline prices — which fell (from a peak in May, 2011 to a bottom in early December, 2011) about $.75 per gallon (as recorded in the chart below provided by the good folks at

Click on graph for larger image.

Coincident with the early part of that price drop we saw consumer activity surge. In fact, each penny of price movement in gasoline prices has the potential to move about $1.3 billion in annual spending from non-discretionary gasoline purchases to more discretionary consumer spending. A $.75 drop in gas prices can therefore free up nearly $100 billion in annualized consumer spending — about 7% of total annual durable goods spending. The flip side of that equation is equally obvious: a $.75 rise in gasoline prices can suck 7% out of consumer durable goods spending. These are non-trivial impacts, and an inversion of the chart might serve as a very good first approximation of the state of consumer discretionary spending. And (like our data) the gas price chart is available with a daily resolution.

Just as our Weighted Composite Index soared last year as gasoline prices dropped, we are now seeing our year-over-year metrics dive as a result of weakening comparisons against much stronger year-ago consumer activities. But our Absolute Demand Index over the past four years can give you a sense of the “absolute” consumer demand (relative to 2005) and remove some of the noise created by year-over-year metrics. In this chart last year’s summer splurge by consumers is clearly visible, as is the general weakening (or re-normalization) during the first half of this year:

Click on graph for larger image.

It is in this context that the “plunge is real” — the surge in consumer activity that we saw in the second half of last year (and as reflected in the GDP numbers for the third and fourth quarters of 2011) has clearly wilted, and comparisons against last year are going to look progressively worse as the latter half of this year unfolds.

(And for those of you alarmed that our “Absolute Demand” index (shown in the chart above) is somehow stuck erroneously at about 85 (on a scale where the average for 2005 = 100), the BEA has also reported that “real” consumer spending on durable goods and new residential construction during the second quarter of 2012 was similarly stuck at 84.1% of the average of those same quarterly spending levels during 2005. You may call this “recovery” whatever you like, but by that BEA measurement per capita consumer discretionary spending is still off some 15% since 2005 — a contraction level normally associated with the word “depression.”)

(2) Retail survey numbers are in fact weakening, but the reports have not yet been alarming enough to garner sensational headlines. Part of the problem is the nature of the reports themselves: they report changes in “nominal sales” on a “same store” basis — i.e., gross sales transactions at retail units that have been open for at least a year. For that reason they carry two biases: (a) they don’t correct for price inflation, and (b) the “same store” methodology introduces a survivor bias when the number of retail outlets are contracting. A clear example of the latter is the case of Borders bookstores, which have simply disappeared from the comparisons against the prior year’s sales (because they are no longer open). In fact, any formerly loyal Borders customers that have been forced to shop at Barnes and Noble will now show up as a net sales growth for retail sales reporting purposes — even if they are actually buying fewer books, but merely now at stores which still happen to be open.

Our numbers, in contrast, are for quantities of like-kind goods (not transacted dollars) and are therefore inflation neutral. Similarly, we are measuring consumer initiated demand regardless of merchant, and therefore we are not subject to merchant “survivor bias.” Furthermore, our data is essentially per-capita, since we use same-shopper metrics to normalize for the inexorable growth of web commerce. And it is axiomatic that US consumers actually “feel” and react to the economy (and presumably vote) on that same per-capita basis.

Additionally, our data is daily and we generally track only discretionary durable goods. Thus our data is both more timely and it measures a particularly volatile part of the consumer economy. Thus our signal will both lead and be amplified relative to surveys that include non-discretionary items (which are generally 90% or more of household budgets).

(Please note that this amplification means that if we were to record a 10% drop in real per-capita consumer discretionary spending on durable goods (excluding residential construction), that drop might result in only a 4% decrease in total consumer goods spending, a 1.4% drop in overall consumer expenditures (the BEA’s Personal Consumption Expenditures “PCE” number) and a 1% contraction in the headline GDP.)

(3) The answer to the “What on earth is going on?” may be simpler than you might expect. We believe that:

— Household disposable income has not grown in any meaningful way for a decade. Any observed changes in long term aggregate consumer spending have been the result of changes in household leverage — which grew phenomenally between 1992 and 2007, and has been contracting (for the most part) ever since. The only exception to the recent general contraction of household leverage has been in student loans — which do not generate substantial discretionary spending, and unfortunately only hamper such spending in newly formed households.

— Over shorter time frames consumers can adjust their discretionary spending for any number of reasons — including gasoline price changes, changing perceptions of the direction of the macro economy, and even short-term budgetary self-medication to relieve “frugality fatigue” (particularly during the holiday season).

We believe that most consumers have a perfectly good grasp of the “direction of the macro economy” — despite happy numbers from the Administration about “X” thousands of new jobs being created each month. We believe that most consumers sense a jobs situation at least as dismal as un-“adjusted” U-6 statistics (where unemployment has risen 1.1% since just last April to 15.2% in July), and they fully understand the persistent reality of such terms as “part time for economic reasons,” “no longer in the workforce” or “long term discouraged workers” — since those terms likely include someone they personally know. The statisticians can stop counting inconvenient truths, but family and friends generally don’t.

Furthermore, we believe that the right hand end of the chart could impact short term consumer behavior in the coming months.

But most of all, we can’t help but think that consumers’ perceptions of the direction of the macro economy are being shaped by a nearly non-stop barrage of TV ads highlighting the failure of the economy to improve over the past four years. We believe the economy should be the primary focus for the Presidential campaigns, but we also appreciate the “collateral damage” that such such a focus will undoubtedly have on the spending psyche of the public.

For all of those reasons we don’t expect any of our charts to reverse course prior to November 6th.

Updated Charts

The drop-off that you can see in the daily version of our Weighted Composite Index is even more spectacular when it is viewed over the past 52 weeks:

and the past 48 months gives some additional perspective to the sudden rise in consumer demand last summer, along with its more recent drop:

When we look for the sources of the downturn, we generally turn to the “usual suspects”:

Even as home loans have dropped to almost 10% below (already depressed) year-ago levels, demand for new rental housing has also plunged:

And the demand for refinancing has also started to “weaken” (if a mere 60% above prior year is somehow weak):

The carry-on from the direct housing numbers is directly felt in the appliances area:

As well as our broader chart for all things “home improvement”:

And our measurements of demand for our Home Improvement retailers shows almost exactly the same pattern:

But that particular tailing-off is not unique to Home Improvement retailers. Notice the same phenomenon in the Discount Retailers:

And it is not just Discounters that are showing that trend. Consider the case of the major chain Department Stores:

And Autos have experienced the same pattern recently. If we look at only the car brands that cater to the “economy” market:

And notice particularly the Korean automakers (which have now fallen to year-over-year contraction for the first time in two years):

By the way, F. Scott Fitzgerald’s observation (that the 1% live differently from the rest of us) seems to hold true for Ferraris et al:

Along those same lines, one of our recurring themes has been the disconnect between Wall Street and “Main Street.” The “recovery” so evident in the earnings of the large cap US equities has not been fully shared by the small businesses, self proprietors or workers on Main Street. And it is on Main Street, after all, that most consumers informally and instinctively gather their intelligence about the state of the economy. They may be naive about the workings of the Federal Reserve and the primary dealers, but they are neither idiots or rubes when observing what is happening in their own back yards.

If the Federal Reserve wanted to reflate the economy through QE operations, consumers aren’t buying. If the Federal Reserve wanted to reflate retail investor interest in the equity markets through QE operations, again consumers interest was fleeting:

Or, if the Federal Reserve and the primary dealers want to reflate the equity market’s “wealth effect” (or more perversely the pool of “greater fools”), yet again consumers aren’t currently buying:

The retail investor still senses that things are not quite right on Main Street, and that the party on Wall Street has either been a tad premature or a well baited trap. Markets flourish in an environment with both transparency and trust. US consumers will sense neither so long as there is a persistent disconnect between Wall Street and what they see happening along their own stretch of “Main Street.”

Related Articles

Analysis articles by Rick Davis

Analysis articles on retail sales