I’m fascinated with financial market history because learning about prior cycles gives you a sense about how some things never change while others are in a constant state of flux. It’s both always and never different this time.
This week I read a number of thought-provoking pieces that discussed market history in terms of where we stand today and what it means for market returns going forward.
The first was from Robert Shiller, writing in the New York Times, about his famous Shiller CAPE ratio and how it’s signaling caution on stock valuations. Shiller has been saying this for a number of years and he’ll probably be right eventually but the history of his indicator is what stood out to me in this piece:
What is CAPE, or the cyclically adjusted price-earnings ratio, exactly? Bear with me. This is a bit technical: It is real, or inflation-adjusted, stock price divided by a 10-year average of real earnings. It is usually measured using the price and earnings of the Standard & Poor’s 500-stock index, adjusted for inflation with the Consumer Price Index. In 1988, John Y. Campbell (now at Harvard) and I showed in a joint article that such a ratio has, since 1881, forecast returns somewhat well in the stock market. That “somewhat” is important because the ratio has its limits as a forecasting tool.
Benjamin Graham talked about using long-term averages of earnings to value stocks way back in the 1930s but Shiller’s CAPE ratio that everyone quotes so often going back to 1881 has only been around since 1988. It’s not like investors have always been aware of the long-term average CAPE ratio. It’s still a relatively new phenomenon that no one really talked about until the late-1990s.
I also came across a piece by the Chicago business school about the origin of the first index fund that drove home the point about how far historical data and technology have come over the years. The following is about a group of researchers who were trying to piece together historical stock market data in the 1960s:
It is hard to conceive of just how little data and few computing resources were available. First, Professor Lorie’s CRSP database, which at its inception had charted the rising and falling stock prices of 506 companies from 1925 to the current day, was brand new. (The team at Wells Fargo was its first outside user.) Other than that, analysts had virtually nothing to work with to begin building their database of information. And without a database, they could never hope to create a system to predict market behavior and manage funds on those predictions.
Second, Wells Fargo didn’t have a single machine capable of running FORTRAN—the analytic language of the time.
“We were able to buy time on the 7094 II at Standard Oil for $500 an hour,” says Wagner. “We were hiring college graduate programmers for $475 a month! This was the largest machine in San Francisco. Quite literally, it had far less capacity than today’s iPhone, but it was room-sized.”
Thirdly, analytic packages didn’t exist.
“When I wanted to run a regression analysis, I had to program a step-wise regression.”
Finally, there was no university that offered a major in quantitative finance.
“Almost all the practitioners had operations research or engineering backgrounds,” says Wagner, who had graduated from Stanford. “Hell, hardly anyone graduated from college even knowing how to program a computer.”
Our knowledge of market history and the collective experience of market participants probably doesn’t get enough credit in shaping how we view the financial markets going forward. Certain key lessons are learned over and over again by every new generation but others seem to build on top of one another.
As markets become more mature, it’s bound to affect performance in ways that are worth considering. William Bernstein called this the paradox of wealth in a research piece in the Financial Analysts Journal:
Far from being the investor’s friends, rapid technological advancement and the attendant wealth it produces are a triple-barreled destroyer of returns by (1) increasing societal wealth (through increased industrial productivity) and hence decreasing the cost of capital by decreasing impatience, adjusting pricing factors, and/or increasing the supply of capital; (2) encouraging enthusiasm among, and capital flows from, gullible investors; and (3) diluting shares as a result of the increase in share issuance required to capitalize new forms of technology and rapidly growing or rebuilding economies.
The idea here is that we are now a more prosperous country with longer life expectancies, greater comforts, and improved technologies. This was bound to have an effect on interest rates and valuations. Why should risk-less assets earn a decent return in such a scenario?
Pseudonymous blogger Jesse Livermore tackled this topic in a different way this week but he also touched on how technology has sent costs falling dramatically in recent years:
In the year 1950, the average front load on a mutual fund was 8%, with another 1% annual advisory fee added in. Today, given the option of easy indexing, investors can get convenient, well-diversified exposure to many more stocks than would have been in a mutual fund in 1950, all for 0%. This significant reduction in the cost of diversification warrants a reduction in the excess return that stocks are priced to deliver, particularly over safe assets like government securities that don’t need to be diversified. Let’s suppose with all factors included, the elimination of historical diversification costs ends up being worth 2% per year in annual return. Parity would then suggest that stocks should offer a 2% excess return over government bonds, not the historical 4%. Their valuations would have a basis to rise accordingly.
I touched on this in a previous piece but it’s worth revisiting. Prices and valuations are higher while interest rates and dividend yields are lower than they were in the past which would suggest lower expected future returns from current levels. But the 10% historical stock market returns people always cite are on a gross basis. How many investors actually earned 10% stock market returns after accounting for costs?
For most of that history trading costs, taxes, bid-ask spreads and fund fees were materially higher. So while gross returns should be lower when the cost of capital is lower and valuations are higher, on a net basis it’s quite possible that future returns won’t be much lower than the net returns earned in the past. It’s not what you earn, but what you keep that matters.
Some people are beginning to suggest that these ideas mean we’re reaching a permanent plateau that will see valuations remain higher than average for the foreseeable future (and thus lower returns). I suppose anything is possible but Bernstein reminds us that opportunities, though fleeting, will still remain:
At some point in the next few decades, investors will almost certainly have opportunities, given adequate fortitude and cash, to purchase securities at near historically low prices, but it seems likely that these windows will be more fleeting than in the past.
The reason for this is simple — all of the knowledge and technological advances in the world are never enough to change behavior.
Whatever happens, it would be prudent for investors to plan on lower future returns than the history books show. But if you’re able to behave, avoid panicking when market disruptions occur, diversify globally and keep your costs in check, you should still be able to earn decent returns on your money.
How Our Memories Shape Market Cycles
Here’s what else I’ve been reading this week:
- How to create your own equity indexed annuity (A Teachable Moment)
- Your brain wasn’t built to handle reality (Bloomberg)
- I hope this doesn’t describe you (Above the Market)
- When saving and investing really matter (Of Dollars and Data)
- Overcoming panic paralysis with your investments (Realsmartica)
- In praise of a nomadic lifestyle (WSJ)
- The state of wealth management in 2017 (Reformed Broker)
- What we said when the world changed (Collaborative Fund)
- The ongoing challenge of content creation & curation (Abnormal Returns)
- Podcast: Robert Cialdini on how persuasion works in business, life and politics (FT Alphachat)