Click for a printable version

Wide valuation spreads and the resolution of uncertainties have generally preceded extended periods of value outperformance.

Examining Value Cycles

Value cycles tend to be driven by investor expectations and uncertainties which cause valuations to bifurcate into two distinct groups: one that reflects a favorable outlook embraced by conventional wisdom (e.g., dot com stocks in the late 1990's) and one that is at odds (e.g., the "old economy" companies during the same period). Valuation spreads widen, creating a substantial opportunity for value investors. Companies that are out of favor then restructure or reposition themselves to adapt and succeed. When the uncertainties around the out of favor abate, value outperforms and spreads narrow.

Today, the obvious issue in markets is "lower for longer" with regard to interest rates, which has created wide valuation spreads. Valuations of companies favored by the "lower for longer" school of thought (e.g., bond proxies such as REITs and dividend paying stocks) benefited as interest rates continued to fall, whereas those viewed as facing an uncertain future in this environment (e.g., banks) have been written off as perpetual underperformers, much like the "old economy" stocks in the late 1990's. Once again, however, these companies are restructuring and adapting. All that remains is for conditions which gave rise to the uncertainty – expectations of falling interest rates – to stabilize. Once that happens, we should be poised for the next pro-value cycle to begin.

In the past, wide spreads have generally led to periods of value outperformance. This article presents the historical evidence of this relationship in developed markets.1

Defining the Cycles

To define value cycles, we compare the performance of an equally weighted portfolio of the cheapest quintile of price-to- book stocks to the performance of the market cap weighted universe. Because of the varying length of cycles and the non- linear path of the market, there is a subjective element to defining a cycle. For the purpose of our analysis, we define turning points in value cycles as occurring when 1) the relative performance is at a peak or trough and 2) the relative performance since the last peak or trough is +/- 1500 basis points. We then define a full value cycle as consisting first of a period when the cheapest quintile underperforms, followed by a period of outperformance.

  1. 1Emerging markets are not part of this analysis due to the limited history available.

Value Outperformance Over the Cycles

Figures 1-3 present the value cycles for the U.S., Europe, and Japan. As we examine value cycles across regions several patterns emerge. First and foremost is that over long periods, value outperformed the broad market. Of the nineteen cycles in developed markets, value outperformedacross 14 out of 19 cycles and the average annualized outperformance is 4.0%. Second is the variability in the length of the cycles, with some as short as a few years and on the other extreme the 'super' value cycle from 1995 through 2007 in the U.S. which lasted 11.5 years. The period of value underperformance is typically shorter than the period of outperformance, with the period of outperformance almost three times as long on average. We also note that value has worked across a range of various economic and market conditions.

  1. 1Cheapest quintile price-to-book of the -1,000 largest U.S. stocks (equal- weighted data). Does not represent specific performance of any Pzena service.
  2. 2Cap-weighted data.
  3. 3Does not yet qualify as a turn in the cycle under our methodology. Source: Sanford C. Bernstein & Co.; Pzena Analysis

  1. 1Cheapest quintile price-to-book of Europe universe -500 largest European stocks (USD-hedged equal-weighted data). Does not represent specific performance of any Pzena service.
  2. 2Cap-weighted data.
  3. Source: Sanford C. Bernstein & Co.; Pzena Analysis

  1. 1Cheapest quintile price-to-book of Japan universe -300 largest Japanese stocks (JPY equal-weighted data). Does not represent specific performance of any Pzena service.
  2. 2Cap-weighted data.
  3. Source: Sanford C. Bernstein & Co.; Pzena Analysis

Value Cycles and Valuation Spreads

If we overlay the periods of outperformance onto the regional valuation spread charts we presented last quarter, the relationship between extreme valuations and the subsequent normalization becomes clear. Figures 4-6 present regional valuation spreads with the shaded bars representing periods of value outperformance.

Looking at the charts, one can see that the periods of value outperformance are generally accompanied by periods of spread contraction, whereas periods of underperformance are generally marked by widening spreads. There is clearly some market behavior that causes spreads to widen and then there is some clear tendency for spreads to return back to normal. And the wider the spreads, the greater the potential relative performance for value as spreads narrow.

But every value cycle doesn't lead to outperformance over the full cycle. During the most recent value cycle in the U.S., we saw a sharp contraction in spreads in early 2009. However, unlike previous cycles where spreads contract from approximately one standard deviation above the historical average to one standard deviation below, spreads stopped contracting prematurely in late 2009. Spreads then started widening on fear of a European crisis which continued with global quantitative easing followed by fear of an economic slowdown in China.Consequently, the relative performance generated across this value cycle was negative.

  1. Universe is -1000 largest U.S. stocks by market capitalization.
  2. Source: Sanford C. Bernstein & Co.; Pzena Analysis.
  3. Data as of September 30, 2016

  1. Universe is -500 Largest European stocks (USD-hedged equal weighted data)
  2. Source: Sanford C. Bernstein & Co.; Pzena Analysis
  3. Data as of September 30, 2016

  1. Universe is -300 largest Japanese Stocks (JPY equal-weighted data)
  2. Source: Sanford C. Bernstein & Co.; Pzena Analysis.
  3. Data as of September 30, 2016.

Value Cycle Interruptions

As we discussed in last quarter's commentary, once spreads widen beyond one standard deviation, the prospective returns of a value strategy to a long-term investor are compelling. But what do we make of an interrupted value cycle? Sometimes, the next crisis unfolds before the prior one was resolved. Japan in the 1990s provides an interesting case study.

In figure 6 we can see the dramatic widening of spreads in 1996 as concerns were mounting about the Japanese banking system. By late 1997 numerous Japanese banks had failed and Japan was facing a domestic financial crisis. Meanwhile, commodities and industrials were hard hit by the Pan-Asian currency crisis. Spreads widened, and value underperformed.

Companies restructured and by early 1998 fear surrounding the Asian financial markets dissipated. Spreads began to come back in line and value outperformed sharply, but not by enough to make up for the underperformance during the dramatic downturn, and annualized relative returns through this period were -3.8%. Before the financial system could completely return to health, dot.com euphoria hit the Japanese market, sending technology valuations soaring and halting the value recovery. It took until the end of the dot.com era for value investors to earn a good return. Those who invested at the beginning of the cycle in 1996 found themselves 3.8% behind the market (annualized) in May of 1999 when spreads prematurely widened and the value cycle was interrupted. When the dot com bubble peaked in late 1999, one was 14.3% (annualized) behind. A pro-value cycle then commenced and combining the two full cycles, value produced annualized excess returns of 9.0%.

A similar phenomenon occurred in Europe when the 1994- 1998 cycle was interrupted by the dot.com era, but the two cycles together produced 6.6% of annualized excess returns. And similar to the U.S., the most recent value cycles have been disappointing because of the massive uncertainty caused by three successive crises – the Global Financial Crisis, the Euro crisis and the slowdown in emerging markets.

Time Is The Friend Of Value

In environments with wide spreads like today, the historical data have been quite compelling; wide valuation spreads and the resolution of uncertainties have generally led to extended periods of value outperformance. Even when the recovery in value is interrupted, as it has been recently in the developed world, staying the course has led to excess returns. Inevitably, the circumstances behind a cycle are exhausted and businesses adjust. Today most of the world is caught up in the same cycle induced by quantitative easing. We expect that eventually this too shall pass, and value will be well positioned to outperform.


Past performance is no guarantee of future results. The historical returns of the specific portfolio securities mentioned in this commentary are not necessarily indicative of their future performance or the performance of any of our current or future investment strategies. The investment return and principal value of an investment will fluctuate over time.

The specific portfolio securities discussed in this commentary were selected for inclusion based on their ability to help you understand our investment process. They do not represent all of the securities purchased, sold or recommended for our client accounts during any particular period, and it should not be assumed that investments in such securities were, or will be, profitable.