We currently estimate the prospective 10-year total return on the S&P 500 at about 4.5% annually, in nominal terms, based on our standard valuation methodology. This may not seem bad, relative to 2% yields on the 10-year Treasury bond, provided that investors actually consider either figure to be an adequate 10-year investment return, and provided that they view 4.5% annual returns as adequate compensation for securities that have several times the volatility of a 10-year Treasury bond (especially when yields are low), and provided that investors ignore the fact that prospective market returns tend to enjoy a significant range over the course of the market cycle, so that « locking in » present prospective returns must necessarily forego any higher prospective return that might be observed in the coming decade. Even given robust growth in GDP and corporate revenues, a move to prospective returns of just 6% at some point in the next two years would likely leave investors with no return (including dividends) in the interim (see Too Little to Lock In).
Wall Street continues to focus on the idea that stocks are « cheap » on the basis of forward price/earnings multiples. I can’t emphasize enough how badly standard P/E metrics are being distorted by record (but reliably cyclical) profit margins, which remain about 50-70% above historical norms. Our attention to profit margins and the use of normalized valuation measures is nothing new, nor is our view that record profit margins have corrupted many widely-followed valuation measures. As I noted in our September 8, 2008 comment Deja Vu (Again), which happened to be a week before Lehman failed and the market collapsed, « Currently, the S&P 500 is trading at about 15 times prior peak earnings, but that multiple is somewhat misleading because those prior peak earnings reflected extremely elevated profit margins on a historical basis. On normalized profit margins, the market’s current valuation remains well above the level established at any prior bear market low, including 2002 (in fact, it is closer to levels established at most historical bull market peaks). Based on our standard methodology, the S&P 500 Index is priced to achieve expected total returns over the coming decade in the range of 4-6% annually. » Present valuations are of course more elevated today than they were before that plunge.
Suffice it to say that every P/E multiple is simply a shorthand for proper discounted cash-flow methods, because there are countless assumptions about growth, margins, return on invested capital and other factors quietly baked inside. Like price-to-forward operating earnings multiples, even our old price-to-peak earnings metric has been rendered misleading due to historically high profit margins. Of course, we knew that was happening even before the credit crisis began, and believe that numerous widely-followed valuation measures remain distorted by record profit margins here.
On the economic front, the recent uptick in new unemployment claims is consistent with the leading economic measures and « unobserved components » estimates that we obtain from the broad economic data here (see the note on extracting economic signals in Do I Feel Lucky?). Indeed, it will be difficult to get the expected flat or negative April employment print if weekly new claims don’t rise toward about 400,000 in the next few weeks. We’ve seen « surprising » weakness in some of the more recent regional surveys such as Empire Manufacturing and Philly Fed. A continuation of that trend would also be informative.
As I noted a few months ago, « examining the past 10 U.S. recessions, it turns out that payroll employment growth was positive in 8 of those 10 recessions in the very month that the recession began. These were not small numbers. The average payroll growth (scaled to the present labor force) translates to 200,000 new jobs in the month of the recession turn, and about 500,000 jobs during the preceding 3-month period. Indeed, of the 80% of these points that were positive, the average rate of payroll growth in the month of the turn was 0.20%, which presently translates to a payroll gain of 264,000 jobs. Notably however, the month following entry into a recession typically featured a sharp dropoff in job growth, with only 30% of those months featuring job gains, and employment losses that work out to about 150,000 jobs based on the present size of the job force. So while robust job creation is no evidence at all that a recession is not directly ahead, a significant negative print on jobs is a fairly useful confirmation of the turning point, provided that leading recession indicators are already in place. » (see Leading Indicators and the Risk of a Blindside Recession).
The upshot is that while I expect a weak April jobs report, we should hesitate to take leading information from what remains largely a short-lagging indicator. We’re already seeing deterioration in economic data, but it remains largely dismissed as noise. An acceleration of economic deterioration as we move toward midyear would be more difficult to ignore. My impression is that investors and analysts don’t recognize that we’ve never seen the ensemble of broad economic drivers and aggregate output (real personal income, real personal consumption, real final sales, global output, real GDP, and even employment growth) jointly as weak as they are now on a year-over-year basis, except in association with recession. All of these measures have negative standardized values here. My guess is that we’ll eventually mark a new recession as beginning in April or May 2012.
Emphatically, however, our concerns about the stock market continue to be independent of these economic expectations, as the hostile investment syndromes we’ve seen in recent months have historically been sufficient to produce very negative market outcomes, on average, even in the absence of economic strains (see Goat Rodeo and An Angry Army of Aunt Minnies). As always, I strongly encourage investors to adhere to their disciplines – including those following a buy-and-hold approach – provided that they have carefully contemplated the full-cycle risk and their ability to stick to their strategy through the worst parts of the investment cycle. What I am adamantly against is the idea that speculators can successfully « game » overvalued, overbought, overbullish markets – particularly in the face of numerous hostile syndromes, near-panic insider selling, speculation in new issues, and broad divergences in market internals, all of which we are now observing.
In the absence of hostile syndromes like we observe today, we generally have more equanimity about market prospects – recognizing the average outcome, but also emphasizing the wide range of individual outcomes associated with a given set of market conditions. The majority of our past market comments are filled with reminders that our expectations are based on average return and risk characteristics, and should not be taken as forecasts about any specific instance. At present, the outcomes that have historically emerged from similar conditions are so uniformly negative that too much equanimity would be misleading.
One way to gauge your speculative exposure is to ask the simple question – what portion of your portfolio do you expect (or even hope) to sell before the next major market downturn ensues? Almost by definition, that portion of your portfolio is speculative in the sense that you do not intend to carry it through the full market cycle, and instead expect to sell it to someone else at a better price before the cycle completes. With respect to those speculative holdings, and when to part with them, my own view is straightforward. Run, don’t walk.
Notes on banking and monetary policy
Banks continue to report seemingly pleasant earnings, as long as one doesn’t look under the hood at the drivers of those reports. Two drivers have been particularly important this quarter. One is the further reduction of reserves against future loan losses, which shows up as a positive contribution to bank earnings. For example, a decline in loan loss reserves was the source of about one-third of the earnings reported by Citigroup. The other driver is something called a « debt valuation adjustment » or DVA. You might recall that as a result of European credit strains last year, investors sold off the bonds of major banks. In the world of bank accounting, the debt was therefore cheaper to retire, so – I am not making this up – the decline in the value of the bonds was booked as earnings. Of course, the value of bank debt has recovered somewhat since then, as investors have set aside concerns about Europe (which we doubt is a good idea). One might expect that since banks booked DVA as a contribution to earnings last year, we would see the opposite effect this quarter. But one would be wrong. As Peter Tchir noted last week, « Morgan Stanley no longer includes DVA in its ‘continuing operations’ headline number. It was a loss of $2 billion this quarter. With 2 billion shares outstanding, that would have wiped out the gain. What bothers me, is that in Q3, when it was a gain of $3 billion, it was part of continuing ops. » It was the same story at Bank of America, prompting one analyst to observe « one-time items are to be ignored when negative, and praised when providing a ‘one-time benefit.' »
Tyler Durden of ZeroHedge has started referring to the Federal Reserve as simply « CTRL+P » – which is brilliant, because it really captures the full intellectual content of Fed policy in recent years. Keep in mind that when the Fed engages in quantitative easing, it purchases Treasury securities and pays for them by creating new base money. From an equilibrium perspective, the U.S. government has financed its deficit in recent years partly by issuing new Treasury debt that was bought by the public, and partly by printing money that is now held by the public (corresponding to the Treasuries bought by the Fed). Of course, the Fed can « unprint » the money, so to speak, by reversing its transactions, and selling those Treasury securities back to the public. But the Fed’s ability to do such massive selling without disruption is unproved, to say the least.
Some have asked why the Fed will ever need to reverse its transactions. Couldn’t the Fed just leave the monetary base out there and perpetually roll the Treasury portfolio forward? The answer depends on what sort of inflation we would like to observe, particularly in the back-half of this decade.
To put some structure on this question, I’ve updated our Liquidity Preference chart (1947-present), which illustrates the close relationship between nominal interest rates and monetary base per dollar of nominal GDP. Currently, the U.S. monetary base amounts to 17 cents per dollar of GDP – a level that is consistent with contained inflation only if short-term (3-month Treasury) yields are held below about 10 basis points. For more on the relationship between the monetary base, interest rates, nominal GDP and inflation, see Sixteen Cents – Pushing the Unstable Limits of Monetary Policy, and Charles Plosser and the 50% Contraction in the Fed’s Balance Sheet.
Think of it this way. The willingness of people to hold a given amount of base money, per dollar of nominal GDP, is intimately tied to the rate of return that they could get on an interest-bearing security. Higher interest rates reduce the demand for zero-interest cash. So if there is upward pressure on interest rates, and the Fed leaves the money supply alone, how do you reach equilibrium? Simple – nominal GDP becomes the adjustment variable. If there’s not enough real GDP growth to absorb the excess base money, prices rise to do the job.
Likewise, expanding the amount of base money per dollar of nominal GDP puts downward pressure on Treasury bill yields and short-term interest rates, but really only if there are no inflationary pressures in the system. Clearly, if inflationary pressures are present (suggesting that the monetary base is already too large), an expansion in the monetary base won’t produce lower interest rates. Rather, it will accelerate those inflationary pressures as nominal GDP is forced to keep up with the monetary base – even if real GDP isn’t growing at all. All hyperinflations are built on this dynamic. That said, it’s worth emphasizing that untethered money growth is invariably a reflection of untethered fiscal deficits (the central bank just buys the government debt and replaces it with money). So significant inflation is ultimately not a monetary phenomenon as much as it is a fiscal one.
In any event, the simple fact is that the Fed can sustain the current size of its balance sheet, without inflationary pressures, only to the extent that people (and banks) are willing to sit on idle, low or zero-interest money balances. In an environment of credit concerns and an increasingly likely implosion of the European banking system (where the fresh leverage taken to pursue the « Sarkozy trade » is now turning into leveraged losses), the short-term willingness to hold idle but « safe » cash balances is quite high. So in the event of additional credit strains, the ability of the Fed to go further out to the right on the Liquidity Preference curve is nearly unconstrained.
The problem is that this policy is inconsistent with any economic environment except one where credit is imploding and the Fed is running the whole show in setting short-term interest rates. As the Fed increases the monetary base, it becomes a greater and greater challenge to reverse those actions in the future. Getting into the position may be as easy as hitting CTRL+P, but getting out of the position promises to be a disruptive nightmare – not to mention the effect that these policies have in distorting financial markets, rewarding reckless lenders, punishing savers, and misallocating capital.
Notably, any exogenous pressure on short term interest rates to even 0.25% (on the 3-month Treasury yield), would effectively require the Fed to move back to the pre-QE2 monetary base in order to forestall incipient inflation pressures. Of course, the Fed could delay that outcome by boosting the interest it pays to the banking system for holding idle reserves. Then again, the Fed already has a balance sheet leveraged more than 50-to-1 against its own capital. So upward interest rate pressure would begin to induce capital losses on the Treasury securities the Fed has accumulated at low yields. Raising interest payments to banks would further strain the Fed’s balance sheet, producing an insolvent Fed while providing a fiscal subsidy to the banking system at taxpayer expense.
Needless to say, I don’t expect that all of this will end very well, but given that the full historical record captures inflation, deflation, recession, expansion, Depression, credit expansion, and credit crisis, we are prepared to respond to a wide range of possible events, without relying on the hope for perpetually high profit margins, endless monetary interventions, absence of major sovereign defaults, stability of the euro-zone, or avoidance of what we view as an oncoming recession. For now, both market and economic evidence remain negative, and we remain accordingly defensive. That will change, but we emphatically view present conditions as being among the most negative subset we’ve observed in the historical record.
As of last week, the Market Climate continued to be characterized by rich valuations and a variety of hostile syndromes (generally related to overvalued, overbought, overbullish conditions, and other variants that capture a general syndrome of « overextended market coupled with a loss of supporting factors »). This places market conditions among the most negative 1% of observations on record, particularly on a 6-18 month horizon, though shorter horizons are clearly negative as well here. Strategic Growth and Strategic International Equity remain tightly hedged. Strategic Dividend Value continues to be about 50% hedged, which is its most defensive position. In Strategic Total Return, we raised our exposure in precious metals shares to about 12% of net assets in response to recent price weakness in that sector. The ratio of gold prices to the XAU is now nearly 10-to-1, which is close to a record high. Historically, gold stocks have been treated as having « insurance » features, and their negative correlation with other stocks was accompanied by premium valuation multiples. At present, many precious metals shares have higher yields than most S&P 500 stocks, and are also significantly depressed relative to gold prices, which suggests a relative margin of defense even if gold prices were to decline substantially. This sector still has substantial volatility, which is why our exposure in terms of net assets is not aggressive (though we would likely increase that exposure on significant economic weakness). Overall, we’re comfortable shifting to a moderately higher exposure in this sector, recognizing that we may observe additional volatility as market conditions change. Strategic Total Return continues to hold a duration of just under 3 years in Treasury securities, and a few percent of assets in utilities and foreign currencies.