Submitted by Lance Roberts of STA Wealth Management,
It was interesting yesterday to watch the media explode in a frenzy of reporting over the "stronger than expected" auto sales. The increase in auto sales to 16.9 million units was certainly a welcome number. However, was it really the "long awaited" sign of economic recovery that it was portrayed to be yesterday?
"These are stunning numbers, especially since the industry is in the midst of some massive, highly publicized recalls," said Anthony Karydakis, chief economic strategist at Miller Tabak in New York.
We would view this as a strong sign of a consumer sector emerging more confident with pivotal positive implications for spending and growth later in the year."
Is this really the case? Do these numbers finally foretell the long awaited economic revival? For that answer, we need to look deeper into the data rather than at a specific data point. As I have stated so often in the past, it is the "trend" of the data that is vastly more important that a single data. Let me give you an example.
The Bureau Of Economic Analysis (BEA) provides a large dataset on motor vehicle sales going back to 1976. The most recent tally of 16.77 million motor vehicle sales for May was a whopping 8.3% increase from May of 2013 which was only 15.48 million units on a seasonally adjusted annualized rate. However, if we look back at the data the comparisons are going to be far tougher in August, November and next March and April as total unit sales all topped 16 million. In other words, May's sales figures were not much better than any of those previous months. In fact, the last three months of reported data have 16.4, 16.04 and 16.77 million units. The increase in sales followed very sluggish sales from December through February as extremely cold winter weather suppressed car shopping activity to around 15 million units per month.
The single data point of 16.8 million units is better than it was in 4th quarter, but about the same as it was in previous months which doesn't tell us much about overall activity. However, what is often missed by the media in the scramble to produce a headline is to provide "context" around the data. By analyzing longer term data trends, "context" is created to provide a deeper understanding of what a single data point actually means.
This series of charts should help provide some "context" to help formulate your own conclusions about the strength of auto sales and the economic recovery.
The chart below shows total motor vehicle sales (seasonally adjusted) back to 1976. The red dashed line is the current level of total sales. The last time that total sales were in a rising trend and reached 16.8 million units was in July, 1998. Current levels are more closely correlated with prior peaks in sales.
As we discussed yesterday with regards to the ISM report, we can remove some of the obfuscation of the "seasonal adjustment" process by using a 12-month moving average to smooth the otherwise volatile non-seasonally adjusted data.
Note: Notice that the early summer months ALWAYS experience a sharp jump following low points of sales in January.
The annualized growth rate of unit sales is on the decline. The red dashed line denotes a 0% change in annualized unit sales. A decline to a -10% annualized rate of change has historically denoted the onset of recessions.
While motor vehicle production has certainly ramped up in recent years, the question is whether or not current sales are keeping up with production. The charts below relate to inventory. High levels of inventory are not a sign of a healthy environment as "inventory gluts" lead to a myriad of other problems. However, if sales are stronger than production, the ratio should be at lower levels which is a sign of a healthy environment.
One of the most important points that is continually missed in the discussion of unit sales is how it relates to the overall population. If 10 units are sold to a population of 100 that is one thing. However, if 10 units are sold to a population of 200, it is an entirely different story. The chart below shows the number of units sold on a per capita basis. I have only used the working age population of 16 years and over which covers all potential domestic drivers.(Note: It is interesting to note that auto sales per capita did not decline much during the recession of 2001. This was due to the fact that most of the economic impact was centered around the decline of the Nasdaq stock market. However, the decline in sales during economically centered recessions was a far different story.
Currently, sales per capita has only recovered back to levels that were only witnessed at the troughs of previous economic recessions. Unfortunately, it appears that the growth in unit sales is entirely due to increases in population, and replacement needs, as opposed to substantially stronger economic health. The historic average for unit replacements has run at about 6% annually. The chart below shows the total increase in population, ages 16 and over, plus unit replacements reduced to 5% as compared to the total increase in unit sales since January 2009.
Lastly, while the media remains focused on total light vehicle sales as a sign of economic recovery; would not "heavy trucks" be a better overall indicator of economic health. If the economy is indeed getting stronger, the increasing level of demand on businesses for production should be reflected by a rise in the demand for trucks. (The red dashed line denotes current level of heavy weight truck sales.)
It is clear that motor vehicle sales have recovered since the financial crisis and along with it a modicum of economic growth. However, it is also important to note that the recovery has been weak at best which leaves little "wiggle" room for an exogenous shock of any kind.
Then there is the question of sustainability. The current level of unit sales has risen from its lows but now appears to be reaching a potential saturation point. At the bottom of the financial crisis, the quantity of sales suggested that the average American would keep their existing vehicle for 25 years. Today, that is no longer the case and much of that excess replacement need is likely filled.
Furthermore, with dealer inventories reaching extreme levels, financial incentives to move cars out of the showroom will continue to become an ever more pressing need. Subprime auto loans are already back in vogue along with little or no money down deals. Since that worked out so well last time, why not do it all over again?
The important point is that "context" is much more useful in the decision making process than the headline driven "knee-jerk" reactions to isolated data points. The motor vehicle data may be telling us a different story than headlines would suggest. If that is the case, the headlines will state that analysts are "shocked" by the "unexpected" outcome.