Retired Or Retiring In Next 15 Years? Better Get Defensive (Part I Of II)

by: James Hickman

Summary

Market timing sensible in certain circumstances - like reducing US equities exposure now.

Always passively invest in public equities and fixed income - not alternatives - but asset allocation still requires active approach.

Financial healing power of "the long-term" is no remedy for max draws in retirement plan homestretch.

Portfolio implications of 3% 10-yr for another decade, 2% US GDP forever.

"Market timing is a loser's game" is a misleading marketing slogan peddled by the long-only mutual fund machine. The mass cash movements in and out of public equity markets that cause market timing failure are rarely driven by disciplined, value-based decisions about asset allocation but rather by emotional investor capitulation to protracted trends at precisely the wrong times. The trite phrase is invariably trotted out when markets are most over-valued and risky - when investors should be selling but rarely are. Now is one of those times.

Recognizing that you should always use low-cost, passive vehicles in certain asset classes and pay for skill in others is not news. But the more important question is: How much should be allocated to each asset class? Asset class and investment strategy exposures, beyond just equities and fixed income, is critical to portfolio diversification and return variation (Brinson, Hood and Beebower - 1986; and Xiong, Ibbotson, Idzorek and Chen - 2010). But can asset classes be timed? The answer is yes.

The professional investment industry has always been animated by failed attempts to systematize alpha generation - to create a better mousetrap for delivering repeatable outperformance of the market and justify higher active management fees. Active managers continued their interminable streak of underperforming the broader markets in 2016. According to S&P Dow Jones Indices' SPIVA US Scorecard for 2016, "Over the 15-year period ending Dec. 2016, 92.15% of large-cap, 95.4% of mid-cap, and 93.21% of small-cap managers trailed their respective benchmarks."

What It Takes to Beat a Passive Benchmark - Asset Classes in Which It Is Possible

The only way to beat a passive benchmark is by forecasting non-consensus outcomes, investing in securities leveraged to those outcomes, and being right. That can happen by a) luck, b) accessing better information than "the market", or c) drawing different and more accurate conclusions on the same information possessed by everyone else. Identifying securities or asset classes expected to behave differently than the market suggests, as reflected in the prevailing asset prices, will produce outperformance as the more accurate, differentiated thesis plays out. Identifying valuation gaps is much harder in markets where information has become commoditized and its dissemination extremely efficient - like public equities and credit markets. It remains a challenge but is accomplished by skilled and resourceful active managers in alternative asset classes for which investment information is harder to access and easier to protect, like venture capital and private equity. Passive investing has long made sense in public equities and fixed income, but manager skill can and has produced consistent alpha in alternative asset classes.

Active managers who frequently beat the market may be rare, but investors should be wary of overplaying the increasingly crowded passive investing strategy in conjunction with the risky bromide "buy-and-hold for the long-term." The latter mantra has historical merit, "over the long-term," and individual investors consistently and significantly lag passive market performance. The latter owes less to management fees and other portfolio management friction costs and more to basic human nature. Thousands of years of evolution favored the survival and procreation of those who recognized safety in numbers. This instinct drives the impulse toward joining the consensus that extrapolates recent trends, regardless of conspicuously high risk in so doing. Result? Buying high (greed/irrational exuberance), selling low (fear/panic) and significantly lagging the market over time.

But the financial healing power of "the long term" may not be helpful to retirees who cannot afford a 50%, 40% or even 30% drawdown in the homestretch of a retirement plan. The S&P 500 has returned 5.0% - 5.5% annually, after inflation and including reinvestment of dividends, over the last 10, 15 and 20 years (see table below). But since the bubble peak in August 2000, market returns, including reinvestment of dividends and after inflation, have compounded at only 2.7%. How did "buy-and-hold for the long-term" work out for post-2000 retirees realizing those returns instead of adjusting equity allocations when valuation was mathematically excessive? Those returns include the last-five-years of 13% annual appreciation. The market now sits at another historically high valuation, propelled by an unprecedented period of easy money. Many public and private pension funds are using 7.5% nominal as the baseline assumption for equity returns, and that could lead to big shortfalls.

Stock Market Returns Over Discrete Historical Periods

As the SPIVA Scorecard demonstrates, return dispersion for long-only public equities and fixed income active managers has been virtually non-existent for a long time, but that of alternative strategies, e.g., Arbitrage, LS Equity and Managed Futures, is significant (see chart below). But using low-cost passive vehicles for exposure to these asset classes does not address the high systematic risk in each.

Dispersion of Individual Mutual Funds' 10-Year Returns by Asset Class (2003 -2012)

Asset Returns by Quartile of 10-Yr Compounded Annual Returns for Individual Funds

Source: underlying data from Morningstar.

The busy table below illustrates the disparate return histories of various "asset classes." It should be noted that asset classes are loosely defined in the table, as strategies or styles within an asset class are dubiously treated as unique asset classes. For instance, small cap and large cap equities are highly correlated with no meaningful, sustainable factor-based separation in respective performance drivers. It is also worth noting that straight averages of annual returns overstate the returns observed in a straight buy-and-hold strategy as percentage losses have a weightier bearing on compounded performance than the same percentage gain. But the larger point holds. In any given year, the asset class weightings have an enormous impact on returns. True diversification can only be achieved through a balanced portfolio across uncorrelated asset classes. That said, who is to say timing asset classes is not just as difficult as timing public equities? Why would paying for active tactical asset allocation make any sense? Two reasons:

  1. As stated, Alternative Asset classes ("Alts") lend themselves to alpha generation by skilled managers, demonstrated by wider active manager return dispersion over full cycles. Hence, performance in these asset classes is not simply a function of movement in the overall asset class. An information advantage is harder to come by in traditional asset classes and when one emerges, it tends to be rapidly disseminated and incorporated into prices.
  2. A sober assessment of absolute values in certain asset classes can surface extreme pricing, on either the long or short side, and support enhanced returns over investment cycles. Commodity prices, real estate, public and even private equity market values are all bound by replacement cost economics - though the duration of cycles is often elusive. Capital is drawn to premium returns, driving them down until capital is withdrawn - the replacement cost cycle. When an asset class is significantly above replacement cost, investors can reasonably expect full-cycle future returns to be lower or even negative. Timing remains a challenge, which is why asset allocation decisions should be premised on a 7-10-year investment horizon. Remember the 2.7% annual return for the S&P 500 since 2000, when the market valuation reached its dearest level in a hundred years. Bond markets also exhibit mean-reversion characteristics and currently have bubble-like downside if interest rates move back to historical levels.

Wide Variation in Asset Class Performance Warrants Some Level of Tactical Allocation

Why Long-term Interest Rates Will Stay Low - Meaning for Equities, Bonds and Asset Allocation

Absolute value should be an anchor for portfolio decisions but should be accompanied by some context on relevant factor trends. Failure to consider this context can lead to value traps. What if the US 10-year yield was to remain at or below 3% for another 10 years? That is our expectation. What if the structural growth rate of the US economy is below 2% compared to 3.2% historically?

Major recessions rooted in the financial system have followed eerily similar deflationary patterns historically - lasting a generation - as measured by depressed long-term rates on sovereign debt. The US financial crisis following the railroad bubble bursting in 1873, the onset of the Great Depression in 1929 and the Japanese financial crisis commencing in 1989, all precipitated greater-than-20-year periods of sub-3% long-term interest rates (see chart below). The US is in year 9 since the so-called Great Recession.

History of Long-Term Yields Post-Financial-Panics

The US stock market has never corrected more than 20% (distinct episodes), peak-to-trough, when the 10-year bond yield was at or below 3.4% (roughly 2% in real terms), except for the collapse ushered in by WWII (see chart below, 10-year yield prior to each correction noted in annotations, e.g., "WWII ~2.5%"). But remember, WWII began ten years into the 1929 financial panic, and the 24 years of sub-3% yields it precipitated. The current yield on the 10-year is 2.35%, and below 0.5% in real terms.

Market Corrections Over 20% Are Uncommon in Low-Rate Environments

Looking at yields another way, stock market investors prior to the late 1950s consistently demanded a dividend yield premium over the US 10-year, assigning far more risk to capital appreciation in the stock market - actually demanding a premium yield to compensate for the risk of capital loss - than investors any time since (see chart below). A simple "black box" system of exiting the stock market whenever the 10-year yield minus the stock market yield ("10-year yield premium") was no lower than negative (0.50%) and buying equities when it was at or below negative (2.0%) would have produced 15 trades between January of 1871 and April 1956, and led to 700% of alpha (1108% cumulative return for the black box and 410% cumulative return for the market). The difference would have been 2180% for the black box if the market were shorted at the sell signals rather than simply exited. Ah, the halcyon days of active management!

That simple model broke down after 1956. The 10-year yield premium over the market yield went positive in the fall of 1958 and did not turn negative again until December 2008. The 10-year yield premium is currently 25 basis points, 3.44% below the average premium between 1958 and 2008.

Premium (Discount) of US 10-Yr Yield v. S&P 500 Dividend Yield

Is it fair to surmise the stock market does not begin to present downside of 20% or more until the 10-year yield reaches 3.5%-4%, and the premium to the stock market yield exceeds 2%? If so, why would any investor not overweight US equities expected to continue benefiting from flight to safety and yield?

Equity market support from protracted anemic bond yields was the basis for our position in the last four years that markets would likely go higher despite high-and-rising systematic risk. But is that systematic risk now near a tipping point? Despite sub-3% long-term rates during the 24-year periods after the respective crises, the compounded annual returns in each of the post-financial-crisis stock markets were -1%, -4%, and 0%, for US-1873, Japan-1989 and US-1929, respectively. The US stock market has grown 5% per year since the pre-Great Recession peak, and 13% in the last 5 years. Marry that ominous historical disconnect with other value-factor concerns:

  • historically over-valued US market (see table below showing 3rd highest cyclically adjusted price-to-earnings ratio "CAPE" in history actually understates overvaluation when interest rates and corporate profit margins are normalized)
  • nearly a full generation of corporate capital allocator wariness about investing in organic growth and a related plateauing in labor productivity
  • corporate profit margins still 40% above historical norms
  • US federal debt as a percentage of GDP is up from 64% to 108% since 2007
  • an unsustainable and worst-in-world US fiscal deficit trajectory (true of advanced economies broadly)
  • sixteen years and counting of US production growth below 2% without a single 4% print
  • deflationary demographic trends
  • high levels of underlying financial risk from a protracted period of market complacency during the central bank's protracted experiment with zero-bound interest rate policies

Stock Market is Historically Overvalued

Investors would be wise to get more defensive. How? We plan to follow up with a detailed asset allocation recommendation.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.