[Editor's Note, September 19: This article has been revised since original publication, as the author has removed some content]
"Be fearful when others are greedy and greedy when others are fearful." - Warren Buffett
I've been quiet here at Seeking Alpha lately and for good reason: the market appears overheated. Unfortunately, there aren't a lot of great "margin of safety" investments in the US right now. The market is not yet in a full "bubble," but the risks for a significant pullback are nevertheless sizable.
There are several problems ahead. Valuations are high with the S&P P/E ratio now at over 19. The economy is still in dismal shape, with true unemployment close to 30-year highs. Meanwhile, terrible policy decisions (e.g. Dodd-Frank, Obamacare, higher taxes), coupled with large budget deficits and increasingly loose Fed policy have created significant risks moving forward.
There are some good investments in this market. I'm still a fan of the insurance sector (KIE), in particular. Even during some of the most inflated US markets in history, you could still find a few good buys. Overall, however, when you consider the high valuations and the high risk factors, I'd view this as a time to be cautious and yes, a bit fearful.
Stock Market Bubbles are Rare; Earnings Crashes are Not
Let's start off taking a look at the overall market valuation. The chart below shows the P/E ratio of the S&P 500 (SPY) from 1900 - 2013, excluding 10 months of data in the last recession (Dec 2008 - Sep 2009). I cut that data out to make the chart more readable, as the P/E ratio briefly spiked over 120 in 2009, due to plunging earnings.
The first thing I want to focus on is a rather surprising take-away: stock market bubbles in the US are rare. It's true there are frequently bubbles on a sector-specific (e.g. housing bubble of last decade) or company-specific level, but full-on market bubbles in the US are not as common as popularly believed.
As evidence, take a look at the historical P/E ratios for the S&P 500 since 1900. If we define a "market bubble" to be an event where the P/E ratio of the market exceeds 25 in a relatively "normal" market (i.e. one where earnings do not plunge), you'll notice that it's not a common occurrence at all. After eliminating 2009, you'll notice very few spikes above a 25x P/E ratio.
The Plunging "E"
Once we do examine the relatively few spikes over 25x in the chart, it also becomes clear that for most of those periods, the market wasn't in a "bubble" per se. Rather the "E" (earnings) part of the equation was rapidly plunging.
Take for instance, 1921 when the S&P 500 P/E briefly pushes above 25. That was actually during the Depression of 1920-1921, one of the worst deflationary events of the past 100 years. Earnings peaked in late 1916 at 30.85, then declined for five straight years until bottoming in 1921 at 3.92.
The next P/E jump over 25 comes in 1933, at the bottom of the Great Depression. The market P/E doesn't hit 25x again for another 59 years in 1992. Once again, we find that this was near the bottom of a recession, with earnings falling 43% from the 1989 peak.
For the first 95 years of data, we have only three examples of the S&P P/E ratio moving above 25x and all three resulted from the "E" (earnings) falling due to a recession or depression, rather than stock prices surging. Finally in 1998, we can make a legit case for a true stock market bubble.
In 1998, the P/E ratio surges up to 34x in 1999 while earnings were still moving upwards. Of course, even in this case, the "bubble" was more pronounced in certain sectors, particularly tech and internet related companies. But at least we have one example of a full-on stock market bubble. However, it's difficult to argue based on this data that there's been another one in the past 113 years. Even on the cusp of the Great Depression, the S&P P/E ratio was only around 17.5x; slightly above the historic norm.
Why do I make this point?
It's a popular myth that the stock market has been in a "bubble" before most of the major crashes of the past century. People believe that greed drove up stock prices to unsustainable levels and view it as a morality tale. In reality, there's virtually no evidence to support this thesis outside of the 90s "Tech Bubble." Rather, what's much more common is that investors underestimate macroeconomic concerns, which rapidly undermine earnings. Suddenly, valuations that were based upon inflated earnings figures look flawed, and this leads to a market crash.
The S&P Valuation is Historically High
That said, valuation is still important, and a higher market valuation implies a thinner margin of safety. For this reason, we should take a look at the historic norms.
If we take every month from the year 1900 onwards and look at the average P/E ratio over that 113-year time frame, we'll come up with about 15.7. Unfortunately, as I mentioned earlier, the data from the last recession is problematic. Earnings fell so dramatically in such a short period of time that the P/E ratio surged over 120 at one point. If we eliminate these ten months of "outliers," the average P/E ratio falls to about 15.1. If we take the median, instead of the average, that falls to 14.6. In other words, in an average market, we should expect the P/E ratio of the S&P 500 to be somewhere around 15.
Earnings plunged in late 2008 and early 2009 and slowly recovered after that. The P/E ratio bottomed out post-recession around mid-2011 at 13.5. The chart below looks at how the market has progressed since then. As of July 2013, the S&P P/E ratio was 19.2.
19.2 is high historically, especially once you eliminate the recession and depression figures (where earnings rapidly fall) that inflate the average. It's well above our average P/E ratio of 15.
Why is the S&P 500 selling at such a high multiple right now? That's a great question that I have no expert answer on. I'd surmise that it's a combination of factors:
(1) Dovish monetary policy,
(2) Low interest rates,
(3) Large fiscal stimulus,
(4) The fact that the market has been moving upwards for 4+ years, and
(5) Excessive optimism in regards to certain sectors of the economy
I'm concerned that investors are underestimating big risks to the economy and we have a market that is valued rather aggressively, with limited room for error.
The Dangers of Perma-Stimulus
What are some of the risks moving forward? Let's start with what I'd call the "perma-stimulus." Too many economists and policymakers seem to employ this line of reasoning:
If the economy is down … WE NEED A STIMULUS!
If the economy is up … it's not up enough; WE NEED A STIMULUS!
If the stimulus fails … it's because we needed MORE STIMULUS!
If the economy is improving … we'll fall back into a recession without MORE STIMULUS!
If inflation is high … unemployment matters more; we need MORE STIMULUS!
Basically, there's no situation that we could possibly find ourselves in that doesn't require stimulus according to many modern economists and policymakers. This type of thinking had become endemic in policy circles and at the Federal Reserve Bank, where Janet Yellen (a perma-stimulus supporter if there ever were one) is considered a leading candidate to replace Ben Bernanke.
Of course, the idea of stimulating the economy has its roots in both Keynesian economic thought (fiscal stimulus), as well as Monetarism (monetary stimulus). Unfortunately, the way the ideas are often interpreted today is a gross bastardization of the way J.M. Keynes and Milton Friedman formulated them.
Keynes believed in using fiscal stimulus during a recession, but also supported the inverse (i.e. fiscal surpluses) in a boom. Friedman viewed tight monetary policy as a cause for the Great Depression, but he also criticized the loose "stimulus" policies of the 1970s, which resulted in way too much inflation.
The simple truth is that we've become addicted to stimulus. When you take a look at fiscal policy over the past 12 years, you'll discover that there's not been a single year that we didn't run a large fiscal deficit. What makes it worse is that those deficits have strayed away from the 0% - 2% of GDP range and have increasingly pushed above 4% of GDP, with budget deficits ranging from 6% - 10% of GDP since 2009.
It's true that the budget deficit is currently falling, but it's also true that we are in Year #4 of an economic recovery and it's still at above-average levels. That's not "Keynesian"; that's just a reckless fiscal policy. What happens if we fall back into recession in 1-2 years? Does the deficit shoot back up to 10%+ of GDP?
Perma-Stimulus and Corporate Profits
One reason the "perma-stimulus" (i.e. constant budget deficits) is important is because it fattens corporate profits. Not surprisingly, corporate profit margins are now at record levels. The chart below shows corporate profits as a % of GDP.
If corporate profits are inflated by government budget deficits, what happens when that ends? J. David Stein of Darby Creek Advisors covers this issue in an article, "Plummeting Federal Deficit Could Lead to Corporate Profit Collapse." The chart below (from Stein) shows the connection between budget deficits and corporate profits.
You can see in the chart how connected corporate profits (the green line) are to government deficits (the red line). If the government budget deficit falls, it's likely that corporate profits will also fall or stagnate.
Of course, the government could continue to run huge deficits, but the likely result is that inflation picks up. It's essentially a lose-lose scenario and given the high valuation of the S&P, leaves me to believe that US stocks, as an asset class, might be somewhat risky right now.
Speaking of inflation…
Inflation Isn't As Low as It Seems
Prior to 1983, CPI including housing prices as one of its criteria. After '83, housing prices were replaced with a concept called Owners' Equivalent Rent ("OER"). You can understand how this could impact CPI by taking a look at the chart below comparing OER to the Case-Shiller 10-City Housing Price Index.
Notice that YOY growth for OER barely budges for most of the period since 1988. Even during the prime housing bubble years, such as 2003, it often only showed 2% YOY gains. For instance in June 2003, OER showed a modest 2.2% increase, while Case-Shiller showed a massive 12.8% gain. While CPI might be reasonable to measure consumer prices, it appears to sometimes be a questionable proxy for overall inflation.
For this reason, I created a housing-adjusted Alternative CPI, which I first showed in my 2011 article, "CPI, Housing Prices, and the Great Stagflation." Once you factor in housing prices, inflation in the early 00s looks very similar to inflation in the 1970s. Likewise, the real economy from 2000 - 2005 looks more like a stagflationary period.
Here we are in 2013 and we're seeing another disconnect between official CPI and housing-adjusted CPI. Case-Shiller numbers are up 11.9% YOY (for June '13), while owners' equivalent rent is up a mere 2.2% for the same period. Using a housing-adjusted CPI measure, our 1.8% CPI inflation suddenly turns into 4.8%. This is the highest reading since the housing bubble.
My point here isn't to argue that my "housing-adjusted CPI" measure is a precise measurement of inflation. Rather, it's to show that many common economic stats are flawed. When economists and policymakers use CPI as a measure for "inflation," it becomes problematic because inflation tends to manifest itself in asset prices (such as housing) first.
By focusing on CPI and other flawed proxies for inflation, the Federal Reserve failed to enact adequate policies during the housing bubble, and they are once again, ignoring lurking dangers. I've even encountered a handful of economists rambling on about the 'dangers of deflation' with housing prices rising almost 12% YOY. Of course, the real "dangers of deflation" are created by loose monetary policies that promote malinvestment, which leads to boom and bust cycles.
While my "Alternative CPI" figures are interesting, we can reach the same conclusions by examining interest rates. Over the past eight months, the average interest rate for a 30-yr fixed-rate mortgage has increased 122 basis points.
While there are other reasons that interest rates could be rising, inflation seems to be the most obvious one.
One final chart looks at the spread between the 30-year fixed rate mortgage ("FRM") and the Federal Funds Rate. The average spread since 1972 is +285 basis points (denoted with a dashed blue line below). Even if we exclude the volatile period from 1972 - 1981, the average spread is still +334 basis points (denoted with a dotted black line).
The spread is now +434 basis points. That is 100+ bps more than both long-term averages, but not completely outside the realm of what we've seen before, with +532 bps being the highest reading in December 2001. Of course, there is a caveat here: while the Federal Funds Rate might be a reasonable measure to use from 1972 to 2009, does it become flawed with the current quantitative easing? Is it possible that due to QE, that the spread between the 30-year FRM and the Federal Funds Rate understates the true gap?
Regardless, there's quite a bit of evidence here that the currently accepted wisdom on inflation is wrong. Inflation seems to be nudging higher, while monetary policy appears very loose given this. This is creating the potential for malinvestment and an eventual bust cycle similar to 2008 / 2009.
Excess Capacity in Lending?
The Federal Reserve has pushed QE, in part, due to the struggles of the banking sector. Five years removed from the start of the financial crisis and the banks are still struggling from an earnings perspective. Then again, maybe the reason bank profits haven't recovered fully is because there's excess lending capacity and this is driving down profit margins. Meanwhile, the Fed's stimulus efforts (coupled with TARP and other programs) are preventing the banking system from normalizing.
Let's think about how free markets typically work. In a free market, when there's excess capacity, prices fall, industries downsize, companies fold, and the market corrects. In our Federally-managed banking system, excess capacity continues to manifest itself, as the Federal government keeps banks alive artificially.
Why do I think there's excess capacity? For one, take a look at some recent research by Raymond James. Raymond James argues that mortgage banking revenue will be under pressure in 2014. What really stood out to me in that article was this chart:
It might seem innocuous, but this is actually one of the scariest economic charts I've seen. Notice that in 2005 mortgage originations were around $775 billion. By September 2008, that number had plunged to $300 billion. Then it slowly rebounded back up to $600 billion and generally stayed in the $400 - $500 billion range for years. In the next twelve months, it's forecasted to fall down as low as $250 billion. That's LOWER than it was during the financial panic in 2008! This isn't that odd historically, however, as it appears we're merely reverting back to pre-housing boom norms.
The problem goes back to free markets vs. controlled markets. When there's too much steel in the market place, steelmakers will produce less. When there's too little demand for lending, however, the Federal government will increase subsidies to the banking sector in order to prevent the banks from failure.
All of that money is being drained from somewhere else in the economy. Likewise, by incentivizing lending where none is needed, we're helping fuel malinvestment. This creates more risk in the real economy and, once again, increases the odds of another 2008 / 2009 crash.
If flawed monetary and fiscal policies are problematic, there's an even larger iceberg looming on the horizon: the implementation of the Patient Accountability and Affordable Care Act (commonly known as "Obamacare"). There are countless articles, books, and studies showcasing the dramatic flaws of Obamacare. I'd recommend reading Forbes blogger Avik Roy for the most complete case against the law.
From an investing perspective, the problem with Obamacare is that it will lead to a few dramatic changes in the economy and increase overall risk. A few examples of the impact:
(1) The US employment market will shift. Low-skilled labor, the type you might find in a Wal-Mart (WMT), Trader Joe's, or Papa John's (PZZA) will now commonly be limited to 30 hours per week, so that the employers can avoid providing overpriced insurance. High-skilled labor, on the other hand, may face the opposite dilemma. Since employers will want to minimize their healthcare exposure, they might decide the easiest way to do this is to maximize the amount of work per employee. In other words, if a company needs 1,000 labor hours of high-skilled labor per week, it might decide it's cheaper to hire 15 people to work 67 hours per week, rather than 25 people to work 40 hours per week.
(2) It will impose a stealth tax on the middle class. Obamacare has numerous direct taxes, but the stealth taxes are more troubling. The most obvious stealth tax will hit the average middle class American, who isn't getting Federal subsidies to purchase insurance. For instance, an American who earns $45,000 per year and has to pay $1,500 or $2,000 more annually for insurance is essentially paying a stealth insurance tax of 3.33% on their income. That's money taken out of another area of the economy, whether it's less investments for a retirement account, or less money spent dining-out.
(3) It will result in more Federal spending. The subsidies will result in increased Federal spending, but the extent of the problem is being dramatically understated in the media. For one, there's the "death spiral" issue that could lead to escalating costs (and hence, escalating subsidies) over the years. But we're also seeing more employers push their employees onto the exchanges (see Trader Joe's) where they'll receive subsidies. All in all, don't be surprised if the Federal budget deficit magically starts to creep upwards again next year as a result of Obamacare.
(4) It will increase uncertainty. Many believe that we'll finally see less uncertainty relating to Obamacare in 2014. I question that wisdom. For one, the law is so completely unworkable, that I have an extremely difficult time envisioning that we won't continue to see major changes.
What gives credence to the idea that uncertainty will remain is the deep-seated dislike of the bill. A recent USA Today poll showed 53% of the country opposed to the law, with only 42% in favor of it. Those would be poor numbers even when the law passed in 2010, but that's horrendous to still see those numbers four years after passage.
The numbers become even uglier once you dig beneath the surface. The 53% opposed includes 41% strongly opposed, with only 26% strongly supporting the law. In other words, the percentage of people who "strongly disapprove" of Obamacare is only 1 percentage point lower than the total support (including weak support) for the law. That's ugly and the numbers could get worse when most consumers realize that their insurance costs are going to skyrocket. Or that they are being taxed for being unable to find reasonably-priced insurance coverage.
No, we shouldn't buy into doomsday predictions and in spite of the negatives, it's doubtful that the entire US economy will collapse. What's more likely is that Obamacare further constrains long-term GDP growth (which is already low), leads to a more mismatched employment market, and takes capital away from investment (and the stock market). Moreover, it's going to increase uncertainty even more.
The Diminishing Margin of Safety
None of this is to say that a recession or a stock market crash is imminent. Like everyone else, I have no idea what the future holds. Perhaps the market crashes next month, or perhaps the market goes on a sustained three year run and the S&P hits 2,000.
What I can do is analyze risk and reward. My reading of all of the above suggests that macroeconomic risk factors are high. Corporate profits are being fattened primarily by stimulus and as inflation becomes a greater threat, there's a greater chance that the Federal government and/or the Federal Reserve will have to pull back. Likewise, the Affordable Care Act is a major risk to the economy that will cause insurance prices to rise, re-orient our employment market, and create additional layers of uncertainty moving forward.
Meanwhile, stock market valuations are high from a historical basis. With the P/E ratio of the S&P 500 near 19.5x, at the same time corporate profits have been rising rapidly, there is a significant risk of "mean reversion." All of this is to say, maybe it's a good time to be a bit fearful.