Tag Archives: Retail Sales

Auto Sales Aren’t Nearly As Strong As Reported

Steve Goldstein recently reported for MarketWatch that closely watched auto sales were leading economists to be more confident in their economic outlook for the rest of the year. To wit:

“Motor vehicle sales reached a seasonally adjusted annual rate of 17.45 million in March, up from 16.57 million in February, according to data from Autodata. That’s the highest reading in three months and represents a recovery from a downbeat start to the year. The MarketWatch-compiled consensus expectation was for a 16.8 million rate.”

Jim O’Sullivan of High-Frequency Economics also made a similar point.

“The data reinforce our view that the slowing in retail sales through February was exaggerated.”

Of course, the March retail sales report out last week was led primarily by a pickup in auto sales as well as gasoline prices. (About 60% of the entire increase in retail sales came from these two components.)

There is an interesting dichotomy currently occurring within the economy. While consumer confidence, as reported by the Census Bureau, soared to some of the highest levels seen since the turn of the century, the hard economic data continues to remain quite weak. As noted by Morgan Stanley just recently:

“Compare the New York Federal Reserve Bank’s current 1Q GDP tracking vs ours – FRBNY is currently tracking 1Q GDP at 3.0% versus us around 1%. The difference is larger than usual and is being driven by the fact that the New York Fed incorporates soft data into its tracking (attempting to tie it econometrically to GDP, a very hard thing to do especially in real-time). Our method translates the incoming hard data into its GDP equivalent. Note that the Atlanta Fed’s GDPNow tracking also focuses on hard data and is currently tracking 1% for 1Q GDP.”

Since then, estimates for Q1-GDP have drifted higher, but the point is the same and the stunning divergence can be seen in the chart attached to that same article which shows the difference between the “hard” and “soft” data specifically. (courtesy Zerohedge)

What is currently expected by those with a more “bullish bias” is the hard data will soon catch up with the soft data. However, optimism may be misplaced if the recent CFO survey is any indication:

The survey generated responses from more than 1,500 chief financial officers, including 469 from North America, and showed that:

  • 67% of those surveyed predicted the U.S. economy would be in recession by the third quarter of 2020,
  • 84% believe a recession will have begun by the first quarter of 2021; and,
  • 38% of respondents predicted a recession by the first quarter of next year.

Event Horizon

Economic cycles do not last indefinitely.

While fiscal and monetary policies can extend cycles by “pulling forward” future consumption, such actions create an eventual “void” that cannot be filled. In fact, there is mounting evidence the “event horizon” may have been reached as seen through the lens of auto sales.

I recently discussed this point with my friend Simon Constable specifically. but the entire Video Cast is excellent.

He is right.

Following the financial crisis the average age of vehicles on the road had gotten fairly extended so a replacement cycle became more likely. This replacement cycle was accelerated when the Obama Administration launched the “cash for clunkers” program which reduced the number of “used” vehicles for sale pushing individuals into new cars.

Combine replacement needs with low interest rates, easy financing, and extended terms and you get a sales cycle as shown below. The problem, as always, is there are only a finite number of people to sell to and once they have bought a car, they aren’t coming back for a while.

This is why auto sales are very cyclical in nature. Not surprisingly, due to the cyclical nature of autos, sales tend to flatten and decline as an economic recession approaches. (Note: When auto sales are reported each month they are annualized. The bar chart shows the over/underestimation of auto sales each month as compared to what actually occurred on an annual basis.)

While the media touts the “jump in auto sales,” it is a far different story when compared to the increase in the population. With total sales only slightly eclipsing the previous record, given the increase in the population, this is not the victory the media wishes to make it sound. In fact, the current level of auto sales on a per capita basis is only back to where near the bottom of recessions with the exception of the “financial crisis.” 

Furthermore, the annual rate of auto sales has slowed dramatically and is approaching levels normally associated with more severe economic weakness.

But slowing auto sales is only one-half of the problem. The problem for automakers is, as always, they continue to produce inventory even though demand is slowing. The cars are then shifted to dealers which have to resort to increasing levels of incentives to get the inventory sold. However, eventually, this is a losing game.

The chart below shows the current 12-month average of the annual rate of change in auto sales as compared to the inverted inventory-sales ratio. As you can see, there is a correlation to rising auto inventories and declines in auto sales. The current data suggests further weakness in auto sales coming.

With more and more dealers offering special incentives to lure buyers as demand slows, we are back to seeing commercials of “employee discounts,” “zero down at signing,” and “additional cash bonuses.” There is a limit to the level of incentives that dealers can provide to move inventory. Eventually, the inventory overhang will be problematic.

Data suggests that is happening now according to the Houston Chronicle:

“New car sales locally and nationally are falling as interest rates have soared and automakers have pulled back on incentives. Interest rates on new financed vehicles averaged 6.4 percent in March, the highest in a decade, according to Edmunds, a California-based automobile data provider.”

Subprime Returns

As we discussed in “People Buy Payments,” Americans are drowning in auto loan debt and changes in interest rates matter…a lot. A new report from the California Public Interest Research Group, or Cal-PIRG, finds the average car loan has increased 75% over the last decade. In all, Americans owe roughly $1.2 trillion in auto loans.

“Americans are taking out more loans, they are taking out much larger loans, and they are taking those loans out for a longer period of time,” said Emily Rusch, executive director for Cal-PIRG.

Given the lack of wage growth, consumers are needing to get payments down to levels where they can afford them. Furthermore, about 1/3rd of the loans are going to individuals with credit scores averaging 550 which carry much higher rates up to 20%. In fact, since 2010, the share of sub-prime Auto ABS origination has come from deep subprime deals which have increased from just 5.1% in 2010 to 32.5% currently. That growth has been augmented by the emergence of new deep sub-prime lenders which are lenders who did not issue loans prior to 2012.

“Recent dire warnings about practices in the subprime car loan industry have drawn comparisons to the 2008 mortgage crisis. In an interview with Bloomberg TV in 2017, investor Steve Eisman—who famously profited off the financial crisis by betting against the market—singled out the auto loan industry. ‘We are in an environment where credit quality has never been this good in anyone’s lifetime, with the one exception of subprime auto,’ said Eisman.

Those lending patterns are now being repeated: Many Wall Street lenders have been pushing auto loans aggressively on subprime borrowers on iffy terms. These loans are then spun up into bonds and sold to investors hungry for auto-loan-backed securities.”

While there has been much touting of the strength of the consumer in recent years, it has been a credit-driven mirage. With income growth weak, debt levels elevated, and rent and health care costs chipping away at disposable incomes, in order to make payments even remotely possible, terms are often stretched to 84 months.

The eventual issue is that since cars are typically turned over every 3-5 years on average, borrowers are typically upside down in their vehicle when it comes time to trade it in. Between the negative equity of their trade-in, along with title, taxes, and license fees, and a hefty dealer profit rolled into the original loan, there is going to be a substantial problem down the road. As noted by Reuters:

Typically, car dealers tack on an amount equal to the negative equity to a loan for the consumers’ next vehicle. To keep the monthly payments stable, the new credit is for a greater length of time. 

Over the course of multiple trade-ins, negative equity accumulates. Moody’s calls this the ‘trade-in treadmill,’ the result of which is ‘increasing lender risk, with larger and larger loss-severity exposure.’ 

To ease consumers’ monthly payments, auto manufacturers could subsidize lenders or increase incentives to reduce purchase prices, though either action would reduce their profits, the report said.”

With more sub-prime auto loans outstanding currently than prior to the financial crisis, defaults rising rapidly and a large majority with negative equity in their vehicles, swapping out to a new car is becoming a near impossible option.

The Federal Reserve recently reported the number of borrowers with auto loans more than 90-days delinquent shot up by 1.5 million in the fourth quarter, reaching a total of 7 million — the highest mark ever in absolute numbers, though not as a percentage of the auto-loan market, which has ballooned over the past seven years.

Consumer pain tends to be a leading indicator for broader economic struggles: An increase in delinquencies could signify waning consumer health, foreshadowing a drop in confidence and an overall spending slowdown, which affects nearly every industry.

As reported by Business Insider:

“Bad consumer loans could also inflict losses on major institutions invested in the loans, which are packaged up and sold as asset-backed securities (ABS). That has the potential, if it gets out of hand, to create systemic risk, as we saw with mortgage-backed loans in the last crisis.

So ugly consumer data is a siren alerting investors, trauma-scarred from the mortgage meltdown, to the next proverbial canary in the coal mine.

The surge in auto defaults has been a source of both confusion and consternation. The Fed called the development surprising, and Goldman Sachs analysts referred to it as “something of a puzzle,” given the broader economic and labor-market strength, and the lack of distress in other consumer credit products, such as mortgages and credit cards.”

While the “cash for clunkers” program by the Obama Administration caused a massive surge in used vehicle prices due to the rapid depletion of inventory at the time, much of that inventory has now been rebuilt. Now, used vehicle prices are dropping sharply, as the market is flooded with off-lease vehicles and consumer demand is weakening.

As noted above, the issue of the trade-in treadmill” is a major issue for auto lenders as default risk continues to increase. Per Moody’s:

The percentage of trade-ins with negative equity is at an all-time high, as is the average dollar amount of that negative equity. Lenders are increasingly faced with the choice of taking on greater risk by rolling negative equity at trade-in into the next vehicle loan. We believe they are increasingly taking this choice, resulting in mounting negative equity with successive new-car purchases.”

And sales of new automobiles isn’t nearly as robust as media headlines purport.

Given the importance of automobiles to the domestic manufacturing sector of the economy, it is becoming apparent the sales of autos to consumers has reached an important inflection point.

The previous recessionary warnings from autos was dismissed until far too late. It is likely not a good idea to dismiss it this time. 

Is There A Problem With The BLS Employment Reports?

Since the end of the financial crisis, economists, analysts, and the Federal Reserve have continued to point to the monthly employment reports as proof of the ongoing economic recovery. Even the White House has jumped on the bandwagon as the President has proudly latched onto the headlines of the “longest stretch of employment gains since the 1990’s.”

Yes, there has definitely been an improvement in the labor market since the financial crisis. I am not arguing that point. The financial markets, investors, and analysts eagerly anticipate the release of the employment report each month while the Federal Reserve has staked its monetary policy actions on them as well.

My problem is the discrepancy between the reports and what is happening in the underlying economy. The chart below shows employment gains from 1985-2000 versus wages and economic growth rates.


As compared to 2000-2016.


See the problem here?

IF employment was indeed growing at the fastest pace since the 1990’s, then wage growth, and by extension, economic growth should be at much stronger levels as well. That has YET to be the case.

Part of the reporting problem that has yet to corrected by the BLS is the continued overstatement of jobs through the “Birth/Death Adjustment” which I addressed recently in greater detail.

“For example, take a look at the first slide below.”


“This chart CLEARLY shows that the number of “Births & Deaths” of businesses since the financial crisis have been on the decline. Yet, each month, when the market gets the jobs report, we see roughly 200k plus jobs.

Included in those reports is an ‘ADJUSTMENT’ by the BEA to account for the number of new businesses (jobs) that were “birthed” (created) during the reporting period. This number has generally ‘added’ jobs to the employment report each month.

The chart below shows the differential in employment gains since 2009 when removing the additions to the monthly employment number though the “Birth/Death” adjustment. Real employment gains would be roughly 4.43 million less if you actually accounted for the LOSS in jobs discussed in the first chart above.”


Think about it this way. IF we were truly experiencing the strongest streak of employment growth since the 1990’s, should we not be witnessing:

  1. Surging wage growth as a 4.9% unemployment rate gives employees pricing power?
  2. Economic growth well above 3% as 4.9% unemployment leads to stronger consumption?
  3. A rise in imports as rising consumption leads to demand for goods.
  4. Falling inventories as sales outpace production.
  5. Rising industrial production as demand for goods increases.

None of those things exist currently.

Unreal Retail

Furthermore, as Jeffrey Snider just addressed, the surging jobs in “retail sales” does not jive with actual retail sales. To wit:

“On the sales side, the last year has been appreciably worse than the dot-com recession and recovery yet employment is moving in the exact opposite direction and with that strange intensity of late. Not only are employment figures showing a more robust hiring scenario now than the late 1990’s, the pace is significantly better than even the housing mania of the middle 2000’s. From April 2003 until August 2005, retail sales clearly accelerated, with the overall average 6.0% during those two and a half years (and the short-term, 6-month MA 7.25% by the end of them). It would make sense, then, that hiring would be sustained and relatively robust, with the BLS suggesting 458k total new retail jobs to go along with those increasingly better sales estimates.”


That means we have worse than dot-com recession levels in terms of sales over the past year from early 2015, but not the contraction in retail employment that went along with them prior. Instead, the BLS suggests that hiring is more robust now than during either the heights of the dot-com or housing bubbles even though sales are nowhere near those periods.”

Something is clearly amiss in what is happening in retail trade. We are likely going to see fairly sharp negative revisions to the data when the BLS eventually gets around to accounting for “retail reality.” 

Profits Drive Employment

Let’s set all of the above data points aside for a moment and just talk about the single most important driver of employment – profits.

Business owners are the single most astute allocators of capital on the planet. Why? Simple. If businesses continually misallocate capital over an extended period of time, they will not be in business for long. If sales are declining – companies tend to reign in hiring as a defense against falling profitability.  If profits are declining due to cost increases, like spiraling healthcare premiums, employment tends to be curtailed. Employment, which is the largest expense for companies, is driven by the rise and fall of profits.

I have smoothed the annual variability of inflation-adjusted corporate profits with real GDP to provide a clearer picture of its relationship to the annual rate of change in employment.


We are currently witnessing what is very likely the peak in employment for the current economic cycle. With layoff announcement rising from virtually every sector of the economy, it will likely not take much more economic weakness to see a rise in unemployment rates.

LMCI Leads

Lastly, the Fed’s on Labor Market Conditions Index (LMCI) tends to lead the overall change in the BLS employment reports. The chart below is a 12-month average of the LMCI as compared to the annual change in employment.


Despite the Fed’s “jawboning” about the strength of the labor market as a reason to “normalize” interest rates, their own indicator likely confirms why they have not done so as of yet. The historic correlation is extremely high and the recent divergence will likely not last long as the LMCI approaches ZERO growth. With economic data continuing to weaken, it will likely not be long before employment reports begin to consistently miss overly optimistic expectations.

It is quite evident there is something amiss about the BLS’ employment reports. Is the disparity simply an anomaly in the seasonal adjustments caused by the depth of the financial crisis? Is there an exceptional and unaccounted for margin of error in the surveys? Or, is it something more intentional by government-related agencies to keep “confidence” elevated as Central Banks globally “paddle like crazy” to keep global economies afloat.

I honestly don’t know those answers. I do know the only question that really matters is:

“Who gets to the end of the race first?”

Lance Roberts


Lance Roberts is a Chief Portfolio Strategist/Economist for Clarity Financial. He is also the host of “The Lance Roberts Show” and Chief Editor of the “Real Investment Advice” website and author of “Real Investment Daily” blog and “Real Investment Report“. Follow Lance on Facebook, Twitter, and Linked-In

3 Things: Fed Problem, Oil – Ain’t 2009, NFIB Un-Optimism

“3 Things” is a weekly post of thoughts I am pondering, usually contrarian, with respect to the markets, economy or portfolio management. Comments and thoughts are always welcome via email,  Twitter, and/or Facebook.

The Fed’s Got A Problem

The most recent employment report sent the financial market pundits abuzz claiming that the economy was on solid footing with no recession in sight. The problem, for anyone willing to actually look at the data, was the underlying data was mostly disappointing.

While the BLS trumpeted 242,000 new jobs in February, wages declined 0.1% and the average workweek fell by 0.2 hours along with aggregate hours worked falling a hefty 0.4%. Furthermore, part-time work soared in February while full-time job growth was mediocre.

But even stranger was that out of the 242,000 jobs, retailing saw a massive jump of 55,000 jobs. This is in a month when stores are not hiring a lot of part-time work to deal with a shopping season. Jeffrey Snider at Alhambra Partners picked up on this by pointing out the differential between actual retail sales and “hiring.”

Starting April 2015, overall retail sales (again, including auto sales) fell below 3% on a 6-month average basis (meaning more than a one-month drop in growth rate) – and have remained closer to 2% than even 3%. In past cycles, that has meant initiation of contraction in retail trade employment and widespread recession. Not this time, however, as the BLS gives us a remarkable +313k gain (including February 2016) over the last 11 months. That equates to an astounding 28.5k per month.”


Pretty amazing.  This is even more confounding when you look at the NFIB’s Small Business Survey which shows real retail sales and expectations of sales both on the decline.


But it is not just in “retail trade” that employment gains are an issue but rather throughout the entirety of the report. The obfuscation of the data is coming from the mathematical “seasonal adjustments” along with the “birth/death” adjustment which are adding jobs that aren’t actually there.

Think about it this way. If the unemployment rate were truly 4.9%, and jobs were actually being created at the fast clip since the 90’s, then labor force participation rates, along with wage growth, should be at similar levels. Right? Uhm....


Of course, the reason that labor force participation rates remain so low is that job creation has failed to exceed the growth rate of the working-age population. With the population growing faster than employment, the number of unemployed living in the shadows continues to swell.


Of course, herein lies the problem for the Federal Reserve, who remains intent on further rate hikes this year, which could lead to a major policy error. The Fed’s own Labor Market Conditions Index, which has now declined into contraction, also suggests that something is very wrong with the employment data over the last several months.


If the employment gains were indeed as strong as the Fed, and the BLS, currently suggest; the labor force participation rate should be rising strongly. This has been the case during every other period in history where employment growth increased. Since the financial crisis, despite employment gains, the labor force participation rate has continued to fall.

This suggests that at some point in the future, we will likely see negative revisions to the employment data showing weaker growth than currently thought. 

The issue for the Fed is by fully committing to hiking interest rates, and promoting the economic recovery meme, changing direction now would lead to a loss of confidence and a more dramatic swoon in the financial markets. Such an event would create the very recession they are trying to avoid.

Oil – This Ain’t 2009

I live in Houston where oil is literally the life-blood of the economy. My wife works in the oil field related sales, my friends work for oil companies as well as my clients who are working and saving for their retirement. I would love nothing more than oil prices to go much higher as it would make my life immeasurably simpler.

Over the last couple of weeks, the rally in oil has gotten the financial media and analysts all stirred up with predictions the “bottom in oil is in” and “$70/bbl oil is on its way by summer.” These views are based on the assumption that the decline in oil prices today is much like what we saw during the “dot.com bust” and during the “financial crisis.”

It’s not.

This morning Liz Ann Sonders from Schwab sent out the following tweet:

This misses the main problem with oil which is not going to be resolved anytime soon. The chart below shows the current supply of oil versus demand. Given that prices over the long-term are a reflection of the supply/demand dynamic, the current problem is quite apparent.


This supply/demand imbalance is not going to be resolved my a mild stabilization in supply but rather a rapid decline in production. However, such a swift decline can not feasibly occur due to the need by oil companies to generate income to continue operations, make debt payments, and generate a profit, albeit at much lower levels, for shareholders. As shown in the chart above, the oversupply of oil was eventually reversed but over a very long period of time.

When the supply/demand imbalance became inverted it paved the way for massive oil price rallies in 2005 and 2009 (highlighted circles in the chart below) as “peak oil” became the prevailing fear.


With oil supplies once again exceeding demand, particularly in a weak global economy, oil companies will once again have to balance reducing supply against maintaining operations and profitability. Therefore, we are once again in position for a long, slow, decline in supply which will lead to sustained lower trading ranges for oil prices. Consequently, this will lead to lower reported profit margins for energy companies for the foreseeable future.

The point here is simple: “This ain’t 2009.”  

Yes, there will terrific trading opportunities for energy-related companies in the future. But first we need to work through the shake out of marginal companies that will file for bankruptcy, consolidation of weaker but stable players and reversions of pipelines back into parent companies. These actions will likely keep action volatile in the sector for a while and there will be significant money lost on speculative bets that go bust.

None of this is a bad thing, it is just the consequence of excess being reverted to a healthier and strong state.

If only Central Bankers understood the same.

The NFIB’s Un-optimistic View

While the media continues to jump on every government skewed statistic to spin into a positive headline, a recent survey of “boots on the ground” companies suggest something else is going on in the economy.

The National Federation of Independent Business recently released their latest survey of small businesses for February. The results were less than optimistic as shown by the sharp decline in their overall outlook over the last couple of months.


As Bill Dunkleberg, Chief Economist for the NFIB, stated:

“Monthly management of monetary policy using data subject to substantial revision is inconsistent with the acknowledged lags in policy and not supportive of real growth which requires more policy consistency. Financial markets, of course, thrive on the variability such policies produce and support a zero-interest-rate policy (ZIRP).

Meanwhile, compared to 2009, consumer interest income is down cumulatively over $3 trillion dollars, an unhappy side effect of Fed policies. Low interest rates are great if they occur in an economy that presents investment opportunities. This happens when the economy is exhibiting solid growth which it has not done in this expansion. The 1983 expansion averaged 650,000 new jobs each quarter compared to 450,000 in this expansion with a labor force 30 percent larger.

NFIB data indicate slow growth in the first quarter following the 1 percent growth rate for the fourth quarter of 2015. The GDPNow forecast from the Atlanta Fed is about 2 percent and the NFIB data basically agree. Owners are very pessimistic about business conditions in the coming months and spending and hiring plans have softened.”

Despite the always optimistic view of the media and Wall Street, businesses that ACTUALLY OPERATE in the economy have a much different view. Not surprisingly, the correlation between NFIB expectations and economic growth are fairly correlated which suggest that the economy is likely weaker than headlines suggest. 


As shown above, the decline in actual and planned sales would also suggest weaker hiring. Not surprisingly, the number of firms increasing employment last quarter also declined which again brings into question the accuracy of recent employment reports. 


Of course, if things were as good economically as we are told by Wall Street and the mainstream media, would the ECB really be needing to drop further into negative interest rate territory and boost QE? 

Just some things to think about.

Lance Roberts


Lance Roberts is a Chief Portfolio Strategist/Economist for Clarity Financial. He is also the host of “The Lance Roberts Show” and Chief Editor of the “Real Investment Advice” website and author of “Real Investment Daily” blog and “Real Investment Report“. Follow Lance on Facebook, Twitter, and Linked-In