I am the Kester and Brynes Professor at Columbia Business School and a Chazen Senior Scholar at the Jerome A. Chazen Institute for Global Business.
ESG advocates and skeptics have strong, but often contrary views, about the merits of long-term investing. As you would expect, the reality is nuanced. For some, active long-term investing in ESG companies is the ultimate ESG play. But most long-term active funds struggle to beat indexes over 10 years, especially after expense ratios and sales loads are factored in. Firms, especially the small ones owned by such long termers, earn alphas over indexes in the long run.
Perhaps newly energized by the rise of ESG investing, increasing discussion and disagreement about the relative merits of "long term investing" have become more common. The pro-ESG side argues that ESG is good for business in the long term, while critics argue that this reliance on long term investing is a way for ESG advocates to duck difficult questions about ESG's underperformance in the short run. In particular, they assert that (1) the distinction between long-term and short-term investing is artificial; or (2) there is no intrinsic financial benefit in investing for the long term, compared to alternatives. Some simply ask if the long-term investor makes a return high enough return to reflect the time the investment is held.
My questions is more basic: 1) do investors in fact invest for the long term and 2) does long-term investing even work? Do more investors talk the long-run game than practice it ("long washing")? Do the large indexers, who claim to be universal owners, vote and act as they though they are long-term owners?
Before we get started, an impatient reader would want me to define "long term." That is precisely why I wrote this piece. I have not seen rigorous definitions or measures of that term in the financial press. But please wait till section 3.0 for rigorous measures of what "long termism" could mean. Till then, trust me and carry on.
1.0 Difficulties of holding long-term active portfolios
Holding an active portfolio for a long time entails several difficult tradeoffs. The asset manager should ideally really get to know management of the company. To understand a company, you have to understand the firm's fundamentals (and actually read a 10-K, my pet peeve) and go beyond ratings, scores and other such dubious recommendations of financial intermediaries. The long-term active holder has to invest a lot of time and energy in understanding the firm's corporate culture, without which the long termer will never get a grip on the operations and intangible value driving future cash flows of its investees.
On top of that, the long termer hopefully votes to maximize long term firm value in the proxy proposal process, negotiates compensation arrangements for the managers of these long-term funds to encourage long- term thinking and less churn in portfolios. In the process, such a long-term vision hopefully influences the compensation contracts of the CEOs of the firms that these long termers own to discourage short term temptations to goose up stock prices or earnings.
2.0 Isn't active long-term holding the ultimate ESG investment?
I have written earlier that Warren Buffett was against ESG, but looking back, there may be an alternate lens through which to understand his ESG stance. Isn't investing for the long term in reality the best way to internalize ESG concerns, today's topic du jour? That might explain why Warren Buffett, a public skeptic of ESG, is perhaps the truest exponent of the long-term thinking underlying ESG. By investing for the long haul, one has no choice but to take into account stakeholders' interests, such as those of suppliers, employees, and customers, without having to puzzle over ESG scores that claim to measure such a proclivity. What about engagement with management that ESG folks like to talk about? By owning a small set of companies for the long run, aren't you inherently engaged with managements of those companies?
I set out to understand whether such asset managers exist and what their own governance arrangements look like in practice. Note that this is a long journey with many angles. Hence, I intend to present the data I uncovered in parts. In this part, I am simply trying to understand who the long termers are and whether they make money, as in, beat a passive index.
3.0 Who are the active long-term investors?
How does one identify a long-term investor? The brain-dead answer would be to look at how long the asset manager holds the stock of a particular company. But that metric can mislead. What should we do with asset managers that mostly rely on index funds and ETFs? They hold individual companies, sometimes for long periods, but barely have the time and money to engage with company management in a meaningful way. On the other hand, we have activists like Trian and Elliott, who hold the stock for a year or two, but arguably have more impact on managerial practices than funds that hold the stock for many years. I'll come back to these points later.
For now, let's exclude primarily passive vehicles such as ETFs and index funds. Next, exclude funds that are effectively closet indexers. That is, funds that claim to be active but largely mirror what the indexes do. Sticking with active managers, is there a useful summary number that captures their long-term philosophy?
One idea is to consider the average number of quarters for which an asset manager holds its portfolio. The academic literature measures a more sophisticated version of the "duration" of the portfolio as the weighted-average length of time that the fund has held equities in the portfolio over the last five years (weighted by the size of each stock position). That is, if the fund holds a small stake for long, that will increase duration measure less than a large position held for longer. An alternative measure, available only for mutual funds, is the ratio of total buys or total sells divided by the total net assets at the beginning of the year.
Research suggests that the average duration for which a mutual fund holds an individual company in its portfolio is only around 1.3 years, as per Cremers and Pareek (2016)! No wonder funds have no real incentive to monitor management. Neither does management have any incentive to listen to a fund that will be gone by next year. So, who are the real long-term owners?
My soon-to-be-colleague, Kalash Jain, excluded indexers (both real and the closet variety that don't say they index but do) and crunched the numbers by aggregating all the form 13-F holdings of all stocks held at the fund-family level. I also requested that he exclude banks as they tend to show up as long-term holders in the data without much diagnostic value because they are essentially custodians of such equity. He shared the list of the top 10 long-term investors with the longest time-horizons, with an average duration of about 30 quarters, as of 2018Q4 (the last quarter of his sample), which can be seen below. There are some well-known long-term funds, like Putnam and Gabelli. Berkshire Hathaway is in the top 50, but not in the top 10.
Three foundations show up in the top 10 list of long-term owners: the Robert Wood Johnson Foundation (the Johnson and Johnson (JNJ) foundation), the Lilly endowment (Eli Lilly, LLY) and the Hershey trust (holding HSY, the Hershey company). I am not sure the foundations are involved in the governance of the respective companies in which they invest. Having said that, their holdings and outlook span generations not decades.
As a comparative data point, JNJ, LLY, and HSY's dividend adjusted returns over the last 10 years have been 158%, 354% and 191% as per S&P Cap IQ. The 10-year S&P 500 return, adjusted for dividends, as per S&P's CAP IQ, as of March-end is around 216%. An investor, other than a foundation, might have sold off JNJ and Hershey by now for barely treading water.
3.2 Asset Managers
So, the asset managers on this list that deserve attention would be (1) Longview Asset Management LLC (appropriately named?) with average quarters of specific stocks held for 77 quarters (almost 20 years!); (2) Putnam with 55 quarters (almost 16 years); (3) ESL Investments and Gabelli Funds (around 49 quarters); (4) First Quadrant (47 quarters); and (5) Chelsea and Schroder (42 quarters).
4.0 What about indexers?
To understand the above table better, I wanted to understand the average duration over which a stock stayed in the S&P 500 index as a benchmark. Calculating a comparable duration for the S&P 500 index is more complicated, but the portfolio turnover rate reported by the Vanguard S&P 500's ETF ranges between 2%- 4% between 2021 to 2017. Turnover is defined as follows: if a fund had assets of $100 million at the beginning of the year and the fund bought and sold assets worth $100 million during the year, then the funds' turnover would be 100%.
A technical aside that some may want to skip: It is very important to note that this definition of turnover is different from the duration measure discussed earlier. Why? Because the $100 million of assets sold could have been held for a day or 30 years and the turnover measure does not distinguish the former from the latter. But the cruder turnover measure is easy to find on most mutual fund websites and my "scientific" duration measure to identify the long termers is not. So as an exercise let's go with the reported turnover measure. If we go with my unscientific eyeballing, the S&P 500 ETF holds stocks between 25-50 odd years (turnover of 4% to 2%).
5.0 Do the active long-term investors earn long- term alphas?
The central question of this entire debate is whether long-term investing even makes money and is worth the while. Consider the prospectus of Longview asset management, which is incidentally based in Toronto, Canada, not the United States. Longview' s website states that its inception date is 9/30/2011 and since then they have earned return of 445% till 12/31/22. This sounds very impressive, except for one major catch. The Canadian dollar has lot of steam against the US Dollar. The translated USD return works out to around 316% return. The S&P 500, for the same period (9/30/2011 to 12/31/22) has earned 324%, as per CAP IQ. Because Longview is privately run, I could not find their annual report to assess whether they had hedged the CAD-USD exchange rate exposure. Beating the index, it turns out, is hard.
Putnam's Large -Cap Value fund seems to be the Putnam fund with the highest assets under management (approximately $18 billion) in the Putnam family. The large-cap fund reports performance in five year rolling windows and benchmarks itself to the Russell 1000 index. Putnam's large-cap fund has been around since 1998 and one could argue that sheer Darwinism must imply that they are doing something right.
However, their returns, before taxes and expenses, as per their annual report for year ended 10/31/22, for the class R fund with no loads (since 1/21/03), comes to 11.51% per annum and beats the Russell 1000 index, their chosen benchmark (10.30% p.a.). The annual expense ratio for class R fund is around 1.13% as per the referenced annual report. Hence, their returns, post expenses, are likely to be very similar to those of cheap ETF's that tracks the Russell 1000, their chosen benchmark. The fund's reported turnover is 17%, implying roughly a six-year holding period. Their prospectus states that they hold 71 companies. That's still a large number to monitor actively.
Perhaps there are other Putnam funds that hold stocks for longer than six years. I looked at every Putnam fund I could find on Barrons.com that earned at least 10% p.a. over the last 10 years and checked its portfolio turnover. I could not find a fund with a turnover lower than 17%.
The Gabelli family, like Putnam, hosts several funds. The largest Gabelli Fund, the Gabelli Dividend and Income Trust, has assets under management of $2.34 billion. Their 10 year and 15 year annual return, as of 12/31/22 is 9.5% p.a. and 7.06% relative to S&P 500's return of 12.56% and 8.81%. So, the pre-expense returns trail the index. However, their turnover is only 5% indicating that they hold stocks for 20 years. This fund seeks to hold stocks with stable cash flows that pay good dividends. Would a fixed income or bond index be the right benchmark to consider in this case? Having said that, the trust, as per their annual report, owns hundreds of stocks. I don't know whether relationship management with companies is likely to be effective if the fund owns these many companies but one can always argue that they manage their top 10 holdings.
Schroders is reportedly one of the largest investment managers in the UK. Schroders markets several mutual funds to the public. Hence, data is not a concern. However, their comparative advantage seems to lie in international stocks and hence the data for specific funds may not be very comparable with those of the other funds presented here. Their largest fund, in terms of net assets of around $5.2 billion, is their emerging markets fund. Hence, I have chosen to set aside Schroders for this analysis.
5.5 ESL, Chelsea and First Quadrant
ESL investments is a privately held hedge fund based in Connecticut and has little or no public data on returns, investment process, incentive contracts and the like. In fact, I could not find a website for ESL investments! First Quadrant does not seem to have a website either. Chelsea Management has a website but not much by way of usable data.
In sum, it is quite hard to argue, based on my limited search here, that long term investing, at least in terms of funds available to the public, pays off in an alpha sense, over a comparable index held for a similar period, after expense ratios are considered. That is somewhat depressing but not entirely unexpected.
6.0 What about performance in downturns?
One of the arguments usually offered by the long- termers that active long-term investing does much better during downturns, especially when correlations are pejoratively speaking going to minus one! To examine that premise, we often look at the so-called capture ratio. The up (down) capture ratio is calculated by dividing the manager's returns by the returns of the index during the period the index is up (down). So, the up and down capture ratio for the index would be expected to be 100. Morningstar reports Putnam Large Cap Value's up ratio is 90 whereas the down ratio is 72. Gabelli's Dividend and Income up and down ratios are almost symmetric at 104 and 105, as per Morningstar. A down ratio of 72 for Putnam's Large Cap Value fund does seem impressive. The question is the premium, if any, that the market is willing to pay for that insurance.
7.0 What about academic work that claims that active long-term investors earn robust alpha?
The best cited paper making such a claim is Cremers and Pareek (2016). The paper states "only those with patient investment strategies (with holding durations of over two years) on average outperform, over 2% per year." The paper looks at data from 1990 to 2013 and considers five-year intervals of returns data, unlike my 10-year focus. I have not replicated the paper to know whether the results extend to 2022 and to a 10-year window.
There is some suggestive evidence that indexing has recently taken off to a level not seen in the recent past and made the lives of long-term active owners very hard. Reports suggest that, in 2000, there were 88 ETFs holding $20 billion of assets; 20 years later there were nearly 7,000, holding $7.7 trillion. Moreover, the 2% out performance is measured with reference to the Fama French five-factor model as the benchmark. I wanted to also look at results scored against a standard benchmark such as the S&P 500. I am not sure that a typical municipal pension fund can explain an alpha over a five-factor model to its constituents. But I don't want to say more before I have had a chance to replicate and update these findings myself.
8.0 What about Berkshire Hathaway, the canonical long-term investor?
A paper argues that Berkshire's holding periods, over the years 1980 to 2006, are shorter than one might think. In particular, the authors, Hughes, Liu and Zhang document that the median holding period of stocks is barely one year. Moreover, 30% of stocks held by Berkshire are sold within six months. Only 20% are apparently held for more than two years. Of course, this does not reflect the standard turnover measure that mutual funds report. Nor does it reflect the weighted duration measure that Kalash gave me. You could also argue that Buffett sold losers and bought winners. All that is legitimate and fine but nonetheless sobering.
In terms of performance, Berkshire has returned 198% over the last 10 years relative to S&P500's return of 217% over the comparable period. So, even Berkshire does not beat the index!
9.0 Do firms held by long-termers do better?
One can invert this question and ask whether companies held by long-termers do better in the long run. There are two advantages to this approach. First, if the long- termers really engage with management and influence the adoption of long-term executive compensation plans or governance arrangements designed to maximize long term firm value, one would expect such companies to do better. Second, this approach circumvents some of the secrecy associated with so many private funds that are long- term holders but reveal nothing about their practices, except that they have to a file a form 13-F with the SEC if they own a significant chunk of a public company's equity.
Kalash Jain looks at this very question and concludes that the answer is yes. It does appear as though firms held by long-term investors do perform better, with caveats. The outperformance, in his Table 4 and 6, for value weighted portfolios is about 2.5% a year, but this is mostly because such companies are high "quality" ones associated with high profits and low investments. If you are wondering why low investments are associated with quality, research shows that firms that invest a lot tend to underperform in the long run, either because CEOs over- invest and build empires or because their bold bets don't pan out, on average. On top of that, long-term investing appears to work well for the firms where the investments go, but we see alpha only in smaller firms, as opposed to the big names. I am not sure how much of this is selection (long term investors pick good companies) as opposed to a treatment effect (long term investors make companies better). But this is precisely the area that long term active might want to invest in: high quality relatively ignored small to medium stocks.
10.0 What about Generation Investment Management Global Equity Fund?
Bringing the conversation back to ESG, I inevitably get asked about Generation Investment Management, the firm associated with former VP Al Gore, which reportedly handily beat indexes over the previous 10 years. Bloomberg had analyzed the holdings and returns of the firm's Global Equity Fund in one of their recent pieces on Generation. I could not find that data myself. Hence, we will go with Bloomberg's bar chart and my attempt at eyeballing those numbers. Their Global Equity Fund started accepting client money beginning April 2005. Relying on that analysis, I find that $100 invested in Generation on 4/1/2005, the first date reported in the Bloomberg data, would have accumulated to an impressive $329 at the end of 2022, before factoring in fees and sales load charges. The index used by Generation as a benchmark, MSCI World Index, would have grown to only $252 by the end of 2022.
It is well known that MSCI World has under-performed the S&P 500 over the last decade. If one had invested $100 in the S&P 500 on 1/1/2005, that number would have become $367 by 12/31/22, if dividends on the S&P 500 had been reinvested, as per CAP IQ. Hence, even Generation does not appear to beat the S&P 500. This episode, of course, raises a bigger question of what the right benchmark is to evaluate long-term investment, a topic I hope to come back to later.
Pro ESG folks will argue that S&P 500 is the wrong index as Generation's portfolio is global and the S&P 500 is designed to be US centric. The anti ESG crowd will say that they don't have time for such niceties and would rather buy a simple index such as the S&P 500 or the Russell 5000. But Generation has many interesting ideas on investor engagement and ESG that I will revisit in future parts.
11.0 Where does all this leave ESG investors claiming to focus on the long term?
If I had to distill all this work into recommendations for the debate on long-termism and ESG investing, I would say:
(1) Long-term investment is perhaps the ultimate ESG play as you cannot be in business if you disenfranchise your stakeholders. This statement works with caveats, of course, especially if the firm generates negative externalities that are not priced or taxed, even in the long run.
(2) Publicly available mutual funds increasingly struggle to beat indexes over the long run, especially after considering management fees.
(3) Consider investments made by foundations which think in terms of generations not decades.
(4) Consider investments made by private funds after appropriate due diligence given that we know little about what these funds do.
(5) Firms, especially the small ones owned by such long termers, earn alphas over indexes in the long run. This is the area that long term passive might want to focus on.
(6) How will the ESG funds eventually compete with indexes that deliver better returns, with a puny expense ratio of 0.03% as in the Vanguard S&P 500 ETF? Perhaps investors will be willing to set aside 10% of their portfolio to ESG and expect it not to lose too much by way of alpha for the satisfaction of having done something good for the planet or workers. But measuring such impact with authenticity and rigor becomes even more important in that world.
An important "to do" on my list is to examine closely the role played by passive indexers such as BlackRock in terms of long-term investing and ESG. Moreover, I am sure I have missed active funds that hold stocks for the long term and still manage to beat indexes after fees. If yes, please let me know.
In future parts, I want to look at several outstanding questions including (i) what kind of investment process is followed by the long- termers? (ii) what do the incentive contracts of the managers of these long-term funds look like? (iii) what about the ESG stance and voting records of the long- termers? (iv) what do compensation contracts look at the firms where long- termers own a significant portion of the equity? (v) do state pension funds, that claim to be long- termers, invest in active long-term funds or do they mostly index? (vi) how to deal with passive indexing's free riding of price discovery services that active long-term investors provide? and (vii) are certain passives paid too much to be passive? Stay tuned.
Important Information: This communication is marketing material. The views and opinions contained herein are those of the author(s) on this page, and may not necessarily represent views expressed or reflected in other Exclusive Capital communications, strategies or funds. This material is intended to be for information purposes only and is not intended as promotional material in any respect. The material is not intended as an offer or solicitation for the purchase or sale of any financial instrument.