In the last 20 years the U.S. stock market has had two massive drawdowns of at least 45% apiece and one significant correction in excess of 20%. The larger of these events, discussed below, impacted the psyche and collective behavior of equity investors as much for the magnitude of the decline in equity prices as for the conditions that led to the indiscriminate selling of stocks among professionals and individuals alike. Figure 1 illustrates all three declines going back to September 2009 with the Russell 3000 Index as a proxy for U.S. equities.

Figure 1. Russell 3000 Index September 1999 to September 2019.

Price-only. Source: Bloomberg.  

The 57% decline in the Russell 3000 between October 2007 and March 2009—the largest of these three events—was drawn out over such a long period of time, accompanied by so many negative headlines, bankruptcies and government interventions, and resulted in such a deep global economic recession, that its effects still linger in investors’ minds today. The impacts of the historic market dislocations that ocurred between 2007 and 2009 have been well documented, although we’re not sure they have been fully understood, as we have yet to test the market’s new regulatory structure with another large, sustained drawdown (especially in credit markets). The point of today’s newsletter, however, is to highlight one of the consequences that the 2007-2009 drawdown had on financial markets that is discussed less frequently but that could prove to be as consequential as regulatory changes or monetary policy experiments in the last decade: the belief that markets can be viewed as a simplistic trade-off between ‘risk on’ and ‘risk off’ and that, with the right set of tools, the move from ‘risk on’ to ‘risk off’ and back can be anticipated.

For large asset managers and individuals alike, the proliferation of exchange traded funds (ETFs), which by most measures has been enormously beneficial to investors, has significantly facilitated the switch between asset classes during ‘risk on’ and ‘risk off’ moves. In simplistic terms and within the confines of the equity market, over short time periods this has meant moving in and out of cyclical industries and sectors, large or small stocks, dividend payers or growth companies, and so on. However, such swings have been very short-lived as the market has, for the most part, moved only upwards since it bottomed in 2009 (except for the summer of 2011, which we’ll discuss separately). This means ‘risk on’ has been the dominant investor mode for the vast majority of the last ten and a half years creating a sense of complacency in the strategy that has worked best above all others since the beginning of the current bull market: growth stocks.
 

Breaking Down the Bull Market

In order to better understand the drivers of returns since the beginning of the current bull market, we simplified the investment landscape, focusing on traditional ‘risk factors’ or exposures for U.S. equities. We started by assuming a passive investor would have had a limited menu of options among traditional factor benchmarks to choose from at the start of the current bull market. Our exercise also endows the investor with perfect foresight in March 2009, allowing him or her to own a single allocation, uninterrupted, to date. The menu of options includes two size benchmarks, two style benchmarks and four size plus style benchmarks, all of which are listed below:

Benchmark	Benchmark Type
Russell 1000 Index Russell 2000 Index Russell 3000 Growth Index Russell 3000 Value Index Russell 1000 Growth Index Russell 1000 Value Index Russell 2000 Growth Index Russell 2000 Value Index	Size (Large Cap) Size (Small Cap) Style (Growth) Style (Value) Size & Style (Large Cap Growth) Size & Style (Large Cap Value) Size & Style (Small Cap Growth) Size & Style (Small Cap Value)

With perfect foresight in March 2009, the choice between large and small would have been a toss-up, with stocks doing very well across the board, as illustrated in Figure 2. The choice between growth and value would have been anything but a close call, with growth stocks returning one and a half times what value stocks earned in the last ten and a half years, as seen in Figure 3. Lastly, as one may assume from the two comparisons above, a combination of size and style in a single exposure would have favored small cap growth over all other options, with large cap growth a close second. This is illustrated in Figure 4.

Based on cumulative, price-only returns. Source: Bloomberg.  

Parallels Between 2008 and 2011

As mentioned in the opening paragraph, there were two other significant drawdowns in U.S. equities within the last 20 years in addition to the 2007-2009 drop. The first one was the 49% decline between March 2000 and October 2002, following the collapse of (mostly) technology stocks as the ‘tech bubble’ burst. The other one came in the summer of 2011 when the Russell 3000 Index fell by 21% between April and September, as a result of twin crises of confidence in Europe and the U.S., both tied to sovereign debt issues (Greek and Southern European debt and S&P’s downgrade of U.S. bonds). While the 2000-2002 decline was closest to the 2007-2009 drawdown, both in terms of duration and magnitude, it is the 2011 crisis that most resembles the 2007-2009 drawdown, as we explain below.

While equities were not exactly cheap in the fall of 2007, they most definitely were not expensive by historical standards. Equity valuations were not the cause of the decline in stock prices that began in October of that year. Rather, as it became clear during the course of 2007 that problems stemming from the decline in U.S. housing prices would not be limited to mortgage bonds and related areas of the credit market, investors began to question the viability of large financial institutions at the heart not only of global financial markets but also of the U.S. economy itself. As investors withdrew from any form of counterparty risk and collateral demands were made of overextended investors, credit markets froze, and the housing crisis rapidly became a global liquidity crunch. Institutions rushed to sell their most liquid assets to meet demands for collateral and to build capital buffers, which took down the equity market with breathtaking speed. In summary, the 2007-2009 drawdown was a liquidity crisis and not a drawdown in equities caused by excessive valuations.

With memories of the 2008 financial crisis fresh in investors’ minds, 2011 began playing out in almost identical fashion as 2008 until prompt and decisive action by the world’s central banks flooded the market with liquidity, avoiding a repeat of the story. This time the source of anxiety and temporary investor panic was the possibility of Greece defaulting on its sovereign debt, potentially dragging down the Euro with it. Contagion spread rapidly to similarly indebted nations (Ireland, Spain, Portugal and Italy), as investors anticipated the ‘next shoe to drop’ and market liquidity was withdrawn in masse (and no, the downgrade of the U.S.’s credit rating in August did not help investor psychology). However, having learned the lessons of 2008, liquidity was rapidly restored with a massive bond-buying program implemented by the European Central Bank, essentially guaranteeing all of the Eurozone’s sovereign debt. Market normalcy followed and a major crisis was averted. Thus, 2011 like 2008, was also a liquidity crisis. Figures 5, 6 and 7 illustrate how the benchmarks used in our analysis performed from the October 2011 bottom, to date, extending the parallel between the 2008 and 2011 drawdowns.

Based on cumulative, price-only returns. Source: Bloomberg.  

By contrast, the drawdown in stocks between March 2000 and the fall of 2002 was a result of extreme valuations in U.S. technology stocks (we won’t rehash the entire episode here) as investors realigned their expectations for unlimited growth, at any price, for Internet and technology-related companies. As the bubble in valuations of mega cap technology stocks deflated, they took down with them the market capitalization-weighted indexes against which so many investors were benchmarked. Thus, the 2000-2002 drawdown was a valuation crisis. Figures 8, 9 and 10 illustrate the performance of the benchmarks from the market bottom in 2002 through the next market top in October 2007. Notice how the winners during that market cycle were almost the opposite of those that have dominated after the two liquidity crises in 2008 and 2011.

Based on cumulative, price-only returns. Source: Bloomberg.  

Distinguishing Between Liquidity and Valuation Driven Events

The distinction between liquidity-based drawdowns and valuation-based drawdowns matters, not because understanding it will help us predict when the next downturn will come (we have no idea), but rather because it helps us understand what the market will do after such decline comes. Markets and, more specifically, market ‘factors,’ will behave very differently depending on the nature of the next serious drawdown, as is clear from the charts included above. This seems particularly relevant today as investors appear to be reconsidering the notion that value investing might not be dead after all. And lastly, highlighting the differences in market returns following liquidity and valuation-based drawdowns allows us to once again make the case for a disciplined methodology that combines both growth and value (GARP) and that is agnostic to size, which is of course a good description of the methodology of the Barron’s 400 Index (B400).

So, next in our analysis, we took the last 20 years of stock market returns (with the Russell 3000 Index as the benchmark) and looked at how each ‘factor’ performed from the beginning of each recovery to the next market top. We included these four periods:

• March 2009 bottom to April 2011 top (before next drawdown)
• October 2011 bottom to September 16, 2019 (as of this writing)
• October 2002 bottom to October 2007 top (before next drawdown)
• March 2009 bottom to September 16, 2019 (entirety of current bull market as of this writing)

We then measured the performance by ‘factor’ across these four periods and included B400 in each of them, ranking each benchmark per period based on cumulative, trough to top, price returns. Lastly, we averaged the rank of each factor benchmark (including B400) across all four periods. The results appear below.

Let’s begin with size. Figure 11 displays the performance of B400 compared to the size benchmarks following each of the events described above. An investor with perfect foresight in March 2009 would have chosen the small cap benchmark and held it through the start of the 2011 correction, outperforming the large cap benchmark by 46 percentage points (152% to 106%). However, one owning B400 would have trailed small caps by only two percentage points.

Following the 2011 correction, the rankings flipped, with large caps beating small caps by 14 percentage points (174% to 160%). B400 would have once again fallen in the middle, although this time closer to the loser than to the winner.

Following the valuation-based drawdown of the early 2000s, an investor would have done best by owning B400, instead of picking between small and large. Small caps came in second, trailing B400 by 26 points, followed by large caps, which lagged B400 by 77 points.

And lastly, an investor buying at the 2009 bottom would have done best by owning B400 too, followed by small caps and then large caps. Figure 11. illustrates how each benchmark did across each period and, on average, across all periods based on its average ranking for all four periods.

There are two points worth making about the rankings, by the way. First, the rankings are not time-weighted, meaning we did not give more weight to the benchmarks that did better over longer time periods than those who did better over shorter time periods. These are simple rankings based on the trough-to-peak dates we arbitrarily picked for our analysis. Second, the last period listed in Figures 11-13, overlaps the first and second periods. In that respect, those rankings are double-counted in the average rank for all periods.
 

Figure 11. B400 Performance Relative to Size Benchmarks for Select Periods in the Last 20 Years

Event Type Prior to Period		Russell 1000 Index	Barron’s 400 Index	Russell 2000 Index
Liquidity	Mar. 2009 – Apr. 2011	106%	150%	152%
	Rank	3	2	1
Liquidity	Oct. 2011 – Sept. 2019	174%	161%	160%
	Rank	1	2	3
Valuation	Oct. 2002 – Oct. 2007	108%	185%	159%
	Rank	3	1	2
Liquidity	Mar. 2009 – Sep. 2019	351%	382%	361%
	Rank	3	1	2
	Average Rank	2.5	1.5	2

Based on cumulative, price-only returns. Source: Bloomberg.  

The difference in performance across style benchmarks and across all four periods is much greater than the difference across size benchmarks, as may be seen in Figure 12. An investor owning the growth benchmark from the March 2009 bottom to date would have outperformed the value benchmark by a remarkable 149 percentage points. B400 would have, once again, fallen in the middle, trailing growth by 49 points but besting value by 100 points. This represents a major role inversion between growth and value that took place, unsurprisingly, from period three starting in 2002, when value stocks outperformed growth by 30 percentage points (over a shorter time frame of only five years vs. ten and a half for the current bull market).

Figure 12. B400 Performance Relative to Style Benchmarks for Select Periods in the Last 20 Years

Event Type Prior to Period		Russell 3000 Growth Index	Barron’s 400 Index	Russell 3000 Value Index
Liquidity	Mar. 2009 – Apr. 2011	109%%	150%	110%
	Rank	3	1	2
Liquidity	Oct. 2011 – Sept. 2019	214%	161%	136%
	Rank	1	2	3
Valuation	Oct. 2002 – Oct. 2007	96%	185%	126%
	Rank	3	1	2
Liquidity	Mar. 2009 – Sep. 2019	431%	382%	282%
	Rank	1	2	3
	Average Rank	2	1.5	2.5

Based on cumulative, price-only returns. Source: Bloomberg.  

Finally, our analysis looked at the performance of benchmarks that combine size and style across all four periods, which may be seen in Figure 13.

Figure 13. B400 Performance Relative to Size + Style Benchmarks for Select Periods in the Last 20 Years

Event Type Prior to Period		Russell 1000 Growth Index	Russell 1000 Value Index	Barron’s 400 Index	Russell 2000 Growth Index	Russell 2000 Value Index
Liquidity	Mar. 2009 – Apr. 2011	106%	108%	150%	161%	143%
	Rank	5	4	2	1	3
Liquidity	Oct. 2011 – Sept. 2019	216%	136%	161%	192%	130%
	Rank	1	4	3	2	5
Valuation	Oct. 2002 – Oct. 2007	91%	125%	185%	168%	149%
	Rank	5	4	1	2	3
Liquidity	Mar. 2009 – Sep. 2019	431%	281%	382%	437%	293%
	Rank	2	5	3	1	4
	Average Rank	3.25	4.25	2.25	1.5	3.75

Based on cumulative, price-only returns. Source: Bloomberg.  

Most noteworthy in Figure 13 is the consistency in B400’s performance across all time periods. While it did rank highest in one of the periods (2002-2007), it never ranked any lower than third and, on average, it ranked number 2.25 out of five. Compare this to the two extremes of our analysis: large cap growth and small cap value. Unlike what you might expect, they were not always at opposite ends of the return spectrum (although that has been the case for the most recent period between 2011 and 2019). Nevertheless, they ranked 3.25 out of five (large cap growth) and 3.75 out of five (small cap value), on average, across all four periods. Large cap growth, in particular, seems to go from feast to famine, so to speak, at least in the context of these five indexes. Notice how it ranked worst for the period between March 2009 and April 2011 and then ranked first from October 2011 to date. And prior to that, it also ranked last in 2002 to 2007 and then second from March 2009 to date. An investor might spot such fluctuations and benefit handsomely from buying or selling a specific exposure at the right time. Such a market-timing strategy, of course, requires perfect foresight. For the rest of us, there’s always B400.

Semi-Annual Reconstitution and Rebalance

The Barron’s 400 Index completed its most recent semi-annual rebalance last Friday, at the close of trading in U.S. stock markets, as is customary on the third week of every March and September. The Barron’s 400 ETF (NYSE: BFOR), which tracks B400, thus opened for trading this morning with a reconstituted and equally-weighted portfolio, replicating its underlying benchmark. The best breakdown of the rebalance is available here from our friends at ALPS, while additional information on the ETF is available here. For more information on the index, please visit our website, where you may also find the full archive of our B400 Newsletters. Lastly for all of B400’s constituents and their current MarketGrader scores, please visit Barrons.com.

For a free trial to MarketGrader’s research, covering 37,000 companies globally and 1,200 ETFs, click here.