How does Warren Buffett generate alpha?

In a recent paper titled "Buffett's Alpha", its authors put it best when they wrote:

...[T]he secret to Buffett’s success is his preference for cheap, safe, high-quality stocks combined with his consistent use of leverage to magnify returns while surviving the inevitable large absolute and relative drawdowns this entails.

In summary, I highlight the sources of "Buffett Alpha's" per hedge fund AQR and NYU researchers:

1. The combination of all of the following are sources of Buffett's returns, where the sum is greater than the parts. Most too often focus on the first factor, but Buffett's success cannot be replicated by merely focusing on the first factor.

  • Conviction in investment strategy in the Graham and Dodd mold
  • Financial cushion (wherewithal) to withstand rough periods and not be forced or panicked into selling
  • Skill to operate with leverageand its inherent risk over many decades.Related to the third factor, I have also shared my layman's explanation of Buffett's use of leverage, which I call OPM or "Other People's Money You Owe". Buffett's use of OPM is responsible for at least HALF of his returnsAnonymous' answer to If Warren Buffett had to start today, could he still reach his current level of wealth?

2. Buffett's investment strategy, especially with regards to stock selection, is grounded in the core investment principles of Graham and Dodd as laid out in Security Analysis (1934) and Intelligent Investor (1949):

  • Safe stocks (low beta and low volatility)
  • Cheap (value stocks with low P/B ratios)
  • High-quality (stocks that are profitable, stable, growing and w/ high payout ratios; high dividend payout ratios typically denote stable, high current dividend income)
Whether we’re talking about socks or stocks, I like buying quality merchandise when it is marked down.
— Warren Buffett, Berkshire Hathaway Inc., Annual Report 2008
Ben Graham taught me 45 years ago that in investing it is not necessary to do extraordinary things to get extraordinary results.
— Warren Buffet, Berkshire Hathaway Inc., Annual Report 1994

Stocks meeting the above criteria, which also meet the criteria used by Buffett, have done well in general.Buffett applies about 1.6-to-1 leverage financed partly using insurance float with a low financing rate, and leveraging safe stocks can largely explain Buffett’s performance.

3. Additional Key Points:

  • Buffett’s performance is outstanding as the best among all stocks and mutual funds that have existed for at least 30 years.
  • Sharpe ratio of 0.76 might be lower than many investors imagine.  While optimistic asset managers often claim to be able to achieve Sharpe ratios above 1 or 2, long-term investors might do well by setting a realistic performance goal and bracing themselves for the tough periods that even Buffett experienced.
  • Secret to Buffett’s success is his preference for cheap, safe, high-quality stocks combined with his consistent use of leverage to magnify returns while surviving the inevitable large absolute and relative drawdowns this entails. 
  • Stocks with characteristics favored by Buffett have done well in general, Buffett applies about 1.6-to-1 leverage financed partly using insurance float with a low financing rate, and leveraging safe stocks can largely explain Buffett’s performance.

4. Note: I have some slightly different takes on a few items laid out in "Buffett's Alpha", but in my opinion, the analysis is pretty much spot on.

  • By biggest point of contention is that I think that behaviorally, the presence of leverage FORCED or at least helped him adhere to the discipline and core principles of investing in more conservative, one might say, boring stocks over a long period of time, without selling, and having the stomach to weather the storms.
  • Regardless, the most admirable aspect in my opinion, is that he stuck to his guns and his style and strategy, and did not become overly aggressive in his investments, when the cost or terms of his leverage were more friendly than at other times (i.e. when insurance profitably was higher than normal and the insurance market was extremely hard with high premiums). Most would have probably have gotten carried away and chased and strayed from his or her core investment principles. And when the insurance market softened, it would have been very difficult to switch gears and invest much more conservatively.
  • Another area of oversight of the research paper "Buffett's Alpha" is the significance of the following. The authors write,

"Empirically, Berkshire owns 63% private companies on average from 1980 to 2011 w/ remaining 37% invested in public stocks. Reliance on private holdings has been increasing steadily over time, from less than 20% in the early 1980’s to more than 80% in 2011."

Why is this important? Because Berkshire's privately owned companies like BNSF have less mark-to-market volatility than publicly traded stocks. What this enables is the ability to write premiums in insurance with more flexibility in volumes against a more stable asset base and equity cushion because privately owned companies are marked on the books at cost or a valuation that is updated much less frequently. This is a TREMENDOUS point that is overlooked.

Insurance regulators are able to become more comfortable with the premiums being written against the book value of private companies held by Berkshire Hathaway - because much of that valuation is static. This helped to control the costs of the insurance float even better, which in turn maximized the chances of ensuring underwriting profitability and minimizing the cost of capital of float leverage.

If you have two moving variables, insurance premium/liability levels vs. asset values, which are interconnected as the risk of one impacts the other, the fixing or increased stabilization of one of these variables enables you to control the risks of the other better.

  • The authors go into "anomalous" cheap cost of capital from the insurance float, but the reality is most of the P&C insurance sector has historically been barely break-even or worse with regards to underwriting profitability, implying a more expensive cost of capital. They do not go into how he was able to achieve cheaper cost of "float" than the majority of the industry by approaching underwriting risk differently and unconventionally than most P&C insurance carriers.
  • So, what did Buffett do differently than the rest of the P&C sector? I delve into it a little bit here: Anonymous' answer to If Warren Buffett had to start today, could he still reach his current level of wealth? In that post, you can search for the phrases: (1) "managerial mindset that most insurers find impossible to replicate. Take a look at the facing page" (2) "institutional imperative"

Source: AQR/NYU research, "Buffett's Alpha". (Please see below for summary.)


My summary of "Buffett's Alpha":

1. Intro: The Secret Behind the Oracle’s Alpha

a. The overarching question – what’s the secret behind Buffett’s performance?

i. Among all U.S. stocks from 1926 to 2011 that have been traded for more than 30 years, Berkshire Hathaway has highest Sharpe ratio. Berkshire has a higher Sharpe ratio than all U.S. mutual funds that have been around for more than 30 years.

ii. During 1976-2011 period, Berkshire Sharpe ratio = 0.76, nearly double overall stock market.  High ratio shows high average returns, but also significant risk and periods of losses/significant drawdowns.

b. If Sharpe ratio shows very good but not unachievably good – how did Buffett get so successful?

i. Boosted returns with leverage.

ii. Stuck to a good strategy for a very long time period, surviving rough periods without forced fire selling or career change.

iii. Estimated use of leverage of 1.6x, boosting both his risk and excess return in that proportion.

iv. In summary:

  • conviction
  • financial cushion (wherewithal)
  • skill to operate w/ leverage and its inherent risk over many decades

c. Key question: how does Buffett pick stocks to achieve a relatively attractive return stream that can be leveraged?

Several key features of his portfolio:

i. Safe stocks (low beta and low volatility).

ii. Cheap (value stocks with low P/B ratios).

iii. High-quality (stocks that are profitable, stable, growing and w/ high payout ratios).  High dividend payout ratios typically denote stable, high current dividend income.

d. Follow-up question: is Buffett’s alpha explained by the key aforementioned features of his portfolio?  

i. Accounting for general tendency for safe, cheap and high-quality stocks to outperform and controlling for these factors makes Buffett’s alpha “statistically insignificant”.

ii. Buffett’s genius in part is early recognition of the validity of these factors, but in combination with:

applying leverage w/o ever having to fire sale

  • sticking to his principles

e. If Buffett’s returns are separated into public stocks and private companies, it’s apparent that both contribute to performance w/ public stocks’ portfolio performing better.  So question emerges – why does Buffett rely heavily on private companies as well, including insurance/reinsurance businesses?

This structure provides steady source of financing, enabling him to leverage his stock selection ability. 36% of Buffett’s liabilities consist of insurance float w/ an average cost below the T-bill rate.

f. In summary, Buffett’s formula is unique access to leverage invested in safe, cheap, high-quality stocks.


2. Data Sources – see paper


3. Buffett’s Track Record

a. $1 in Berkshire in 1976 is more than $1,500 at end of 2011.

During this time, average annual return of 19.0% in excess of T-bill rate, outperforming market’s average excess return of 6.1%.

b. Berkshire stock entailed more risk, realizing a volatility of 24.9%, higher than market volatility of 15.8%.

However, excess return was high even relative to its risk: 19%/24.9% = 0.76 Sharpe ratio compared to market’s Sharpe ratio of 0.39. Realized market beta of only 0.7.

c. Performance indicates impressive returns associated with some risk.

i. 6/30/98 – 2/29/00, Berkshire lost 44% of its market value while overall stock market gained 32%.

ii. Many fund managers would have had not survived a 76% shortfall to the market, Buffett’s impeccable reputation and unique structure as a corporation allowed him to stay the course and rebound as internet bubble burst.

d. Berkshire in top 3% among all mutual funds and top 7% among all stocks.

i. Berkshire lies in the very best tail of the performance distribution of mutual funds and stocks surviving at least 30 years.

ii. Compared to stocks w/ at least 30-year history from 1926 to 2011, Berkshire has realized highest Sharpe ratio.


4. Berkshire’s Leverage

Assume:

a. Leverage is used to magnify returns, but the key questions are:

i. How much leverage?

ii. What are the sources of leverage?

iii. What are their terms and costs?

b. One can measure Berkshire’s leverage (L) as follows:

i. Leverage:

Leverage equation.png

OR

L = (MV Assets – Cash) / MV Equity

 

ii. Where market value of the total asset is:

OR

MV Assets = BV Assets + (MV Equity – BV Equity)

 

iii. Based on this method, leverage is 1.6 to 1.

To further elaborate per the authors: "Based on this method, we estimate Buffett’s average leverage to be 1.6-to-1. This indicates a non-trivial use of leverage. This magnitude of leverage can help explain why Berkshire realizes a high volatility despite investing in a number of relatively stable businesses. By focusing on total assets to equity, we capture all kinds of liabilities."

c. Per the authors:

Focusing on total assets to equity, captures all kinds of liabilities and Berkshire’s financing arises from a variety of types of liabilities. The two main liabilities are debt and insurance float and, and leverage can be alternatively calculated focusing on these latter two liabilities:

OR

(MV Equity + Debt + Float) / MV Equity

 

Based on this calculation, focusing on debt and insurance float, leverage comes to 1.4 to 1. i.     Interestingly, Berkshire’s 25% stock volatility is 1.4 times higher than the 17% volatility of the portfolio of public stocks (assuming private assets have similar volatility and ignoring diversification effects).

d. However, magnitude of leverage explains only part of Buffett’s outperformance to the market. 1.6x to the market magnifies market’s average excess return to roughly 10%, far short of the 19% average excess return achieved.  

e. Sources of leverage and their terms/costs:

i. Debt rating: Berkshire’s debt is highly rated, enjoying AAA from 1989 to 2009.  Issued first ever negative-coupon security in 2002, a senior note w/ a warrant.

ii. Float: More anomalous cost of leverage stems from its insurance float.  Collecting insurance premiums up front and later paying a diversified set of claims is akin to taking a loan.  Estimated average annual cost of Berkshire’s insurance float is only 2.2%, more than 300bps BELOW the average T-bill rate.  Low-cost insurance and reinsurance businesses have given him a significant advantage in terms of unique access to cheap, term leverage.  (36% of Berkshire liabilities consist of insurance float on average.)

iii. Tax deductions: Berkshire appears to finance part of its CAPEX using tax deductions for accelerated depreciation of PP&E under IRS rules; $28 billion of such deferred tax liabilities at end of 2011.  These tax deductions are similar to taking an interest free loan in that tax savings (loaned by U.S. gov’t) are enjoyed earlier and paid back eventually in same dollar amount w/o interest.

iv. Accounts payable and derivative contract liabilities: This includes put options and CDS.  Berkshire’s sale of derivatives can serve as a source of financing and a source of revenue, as such derivatives tend to be expensive.  As investors unable/unwilling to use leverage will pay a premium for instruments that embed leverage (i.e. options, levered ETFs), Buffett can profit by supplying embedded leverage as he has unique access to stable and cheap financing.


5. Decomposing Berkshire: Public Stocks vs. Private Companies

a. Public stock performance = Buffett’s stock selection ability. Performance of privately held companies also captures Buffett’s success as a manager.

b. Public stock performance is measured using 13F filings to construct a monthly time series of returns.

c. Private stock value is backed into as the residual amount from the market value of total assets less market value of publicly traded stocks and less cash. See MV of Assets above in 4.b.ii.

i. MV Private Stocks = MV Assets – MV Public Stocks – Cash.

ii. Return on Private Stocks (t+1)  = [Risk-free return (t+1) * MV Liabilities (t) + Return on Berkshire Stock (t+1) * MV Berkshire Stock (t) – Return on Public Stocks (t+1) * MV Public Stocks (t) – Risk-free return (t+1) * Cash (t)] / MV Private Stocks (t).

iii. Note that for Berkshire, MV Liabilities = MV Assets – MV Equity.

iv. Estimate of private holdings portfolio includes value market attaches to Buffett himself (since it is based on overall value of Berkshire stock).  This Buffett-specific element may make estimated value/return on private companies noisy.

d. Hence, Berkshire’s excess return to the risk-free rate can be equated to a weighted average return on public and private holdings, multiplied by a leverage factor:

i. Assuming:

w (t) = relative weight on private holdings at time t and L (t) = leverage at time t,

Return on Berkshire Stock (t+1) – Risk-free return (t+1) =

[w (t) * (Return on Private Stock (t+1) – Risk-free return (t+1))  + (1 – w (t)) * (Return on Public Stock (t+1) – Risk-free return (t+1))] / L (t).

ii. w (t) = MV Private Stocks (t) / (MV Private Stocks (t) + MV Public Stocks (t)).

iii. Empirically, Berkshire owns 63% private companies on average from 1980 to 2011 w/ remaining 37% invested in public stocks.  Reliance on private holdings has been increasing steadily over time, from less than 20% in the early 1980’s to more than 80% in 2011.

e. Both Buffett’s public and private positions do well, exceeding the overall stock market in terms of average excess returns, risk and Sharpe ratio.

i. The public holdings have a higher Sharpe ratio than the private stocks, suggesting Buffett’s skill comes mostly from his ability to pick stocks, as opposed to his value-add as a corporate manager.

ii. Berkshire’s overall stock return is FAR above the returns of BOTH the private and public portfolios.  This is because Berkshire is not just a weighted average of the private and public components. It is leveraged, thus magnifying returns.

iii. Also, Berkshire’s Sharpe ratio is higher than those of the public and private components, reflecting the benefits of diversification (and possibly benefits from time-varying leverage and time-varying public/private weights).


6. Buffett’s Alpha and Investment Style: What Type of Stocks?

a. Returns are attributable to Buffett’s stock selection and ability to apply leverage – but how does he pick stocks?

In line with Graham and Dodd's Security Analysis published in 1934:

Whether we’re talking about socks or stocks, I like buying quality merchandise when it is marked down.
— Warren Buffett, Berkshire Hathaway Inc., Annual Report 2008
Ben Graham taught me 45 years ago that in investing it is not necessary to do extraordinary things to get extraordinary results.
— Warren Buffett, Berkshire Hathaway Inc., Annual Report 1994

b. To answer this question, one can look at 4 standard factor exposures and run a regression for each of the excess returns of Berkshire’s stock, its public holdings and private holdings. For each of these returns, the following factors are accounted for:

i. Market exposure: MKT, market return.  Berkshire has a beta less than 1 and a significant alpha.

ii. Size: SMB, small-minus-big.  Strategy of going long small stocks and shorting large stocks.  Berkshire has a negative loading on SMB reflecting a tendency to buy large stocks.

iii. Value: HML, value factor. Strategy of buying high book-to-market stocks while shorting low book-to-market stocks.  Berkshire’s positive loading reflects tendency to buy cheap stocks on a book-to-market basis.

iv. Momentum: UMD, momentum factor. Strategy of buying stocks that have been “up” in the sense of outperforming the market, while shorting stocks that are relatively “down” on the same basis.  Berkshire’s insignificant loading on UMD shows Buffett is not chasing trends in his stock selection.

c. As shown above, these standard 4 factors do not explain much of Buffett’s alpha.  Because it cannot be explained by these 4 factors studied by academics, his success has been attributed to unique skill or “alpha”.

d. However, when controllingfor (1) the Betting Against Beta (BAB) factor of Frazzini and Pedersen (2010) and (2) the Quality Minus Junk (QMJ) factor of Asness, Frazzini and Pedersen (2012b), Berkshire load significantly on both of these factors.

i. Loading on BAB reflects a tendency to buy safe (i.e. low-beta) stocks while shying away from risky (i.e. high-beta) stocks.

ii. Loading on QMJ reflects a tendency to buy high-quality companies that are profitable, growing and paying out dividends.

e. Controlling for these 2 factors drives the alpha of Berkshire’s public stock portfolio down to a significantly INSIGNIFICANT annualized 0.1%, meaning these factors ALMOST COMPLETELY explain the performance of Buffett’s public portfolio.

i. Significant part of the secret behind Buffett’s success is the fact that he buys safe, high-quality, value stocks.

ii. This also explains a large part of Berkshire’s overall stock return and the private part in the sense that their alphas become statistically insignificant, although it is worth noting that the point estimate of Berkshire’s alpha only drops by about half.

iii. While Buffett is known as the ultimate value investor, his focus on safe quality stocks may in fact be at least as important to his performance.


7. A Systematic Buffett Strategy – see paper a.

Per the authors:

Whereas Buffett is known as an active stock picker, we will try to go back to Buffett’s roots and, in the spirit of Graham and Dodd (1934), focus on systematically implemented screen...In summary, if one had applied leverage to a portfolio of safe, high-quality, value stocks consistently over this time period, then one would have achieved a remarkable return, as did Buffett. Of course, he started doing it half a century before we wrote this paper!

8. Conclusion: Lessons from an Alpha Male

a.    Buffett’s performance is outstanding as the best among all stocks and mutual funds that have existed for at least 30 years.

b.    Sharpe ratio of 0.76 might be lower than many investors imagine.  While optimistic asset managers often claim to be able to achieve Sharpe ratios above 1 or 2, long-term investors might do well by setting a realistic performance goal and bracing themselves for the tough periods that even Buffett experienced.

c.    Secret to Buffett’s success is his preference for cheap, safe, high-quality stocks combined with his consistent use of leverage to magnify returns while surviving the inevitable large absolute and relative drawdowns this entails.

d.    Stocks with characteristics favored by Buffett have done well in general, Buffett applies about 1.6-to-1 leverage financed partly using insurance float with a low financing rate, and leveraging safe stocks can largely explain Buffett’s performance.

"Buffet's Alpha" References:

To give where credit it's due, please give thanks to AQR and NYU researchers for their great work on the paper titled "Buffett's Alpha".