Tweedy Browne has long been one of the most respected and iconic names in the mutual fund industry. Morningstar reported that as of the end of 2014, the fund family had more than $10 billion in assets in its four funds. But, over the long term, has the firm’s flagship fund outperformed a passive benchmark?
The firm has a reputation as a value-oriented investor that follows the principles of Benjamin Graham and David Dodd. In fact, in a 1984 speech Warren Buffett gave in honor of the 50th anniversary of the publication of Graham and Dodd’s book, Security Analysis, he identified Tweedy Browne as one of “The Superinvestors of Graham-and-Doddsville.”
Further establishing the fund’s credibility, Tweedy Browne American Value was selected as one of nine “dyed-in-the-wool value investors” by legendary Sequoia Fund Chief Executive Bob Goldfarb in 2006. Goldfarb previously had been asked by Columbia University professor Louis Lowenstein to select a set of funds that best followed the essential edicts of Graham and Dodd.
At the beginning of 2015, the Tweedy Browne Value Fund (TWEBX) carried a four-star rating from Morningstar, which considers it a large-value fund. Given its great reputation, Advisor Perspectives asked me to analyze the fund’s performance. I’ll do that by comparing it to a passively managed alternative, the U.S. Large Cap Value Fund (DFLVX) from Dimensional Fund Advisors (DFA). (Full disclosure: My firm, Buckingham, recommends Dimensional funds in constructing client portfolios.)
By comparing the performance of these two funds, we can determine not only if TWEBX outperformed a well-designed, passively managed U.S. large-value fund, but also if the freedom to “scour the globe” for the best values provided great opportunity.
In other words, did the fund managers’ ability to tactically allocate add or subtract value?
The evidence
Coincidently, for both the Tweedy Browne fund and the DFA fund, the first full year of operation was 1994. So I chose that as the starting point for our analysis. I’ve chosen to break down the analysis into not only the full 21-year period from 1994 through 2014, but also into four sub-periods. Those sub-periods are the last 15 calendar years (2000-2014), the last 10 calendar years (2005-2014), the last five calendar years (2010-2014) and the first full six calendar years (1994-1999). I’ll begin with the earliest period, 1994 through 1999.
The table below shows the annualized returns, standard deviation and Sharpe ratios for both funds from 1994 through 1999.
|
Fund
|
Annualized Return (%)
|
Standard Deviation (%)
|
Sharpe Ratio (%)
|
|
Tweedy Browne Value (TWEBX)
|
17.07
|
7.11
|
0.43
|
|
DFA U.S. Large Value (DFLVX)
|
15.61
|
8.76
|
0.33
|
As you can see, TWEBX not only produced a higher return, but it did so while exhibiting a lower level of volatility. The result was a Sharpe ratio that was 30% higher than the DFA fund.
Results like these are how great reputations are made. But did that success predict future outperformance?
The next table shows the funds’ results from the most recent five calendar years, 2010 through 2014.
|
Fund
|
Annualized Return (%)
|
Standard Deviation (%)
|
Sharpe Ratio (%)
|
|
Tweedy Browne Value (TWEBX)
|
9.85
|
5.75
|
0.44
|
|
DFA U.S. Large Value (DFLVX)
|
17.02
|
9.18
|
0.48
|
In the most recent five-year period, TWEBX’s return was 7.17 percentage points lower than that of DFLVX. And DFLVX’s Sharpe ratio was about 10% higher in relative terms.
The next table shows the results for the most recent 10-year period, 2005 through 2014. Note that it was early in this period (2006) when the fund’s name was changed from Tweedy Browne American Value to Tweedy Browne Value.
|
Fund
|
Annualized Return (%)
|
Standard Deviation (%)
|
Sharpe Ratio (%)
|
|
Tweedy Browne Value (TWEBX)
|
5.90
|
14.56
|
0.37
|
|
DFA U.S. Large Value (DFLVX)
|
8.10
|
22.59
|
0.41
|
In the most recent 10-year period, TWEBX’s return was 2.2 percentage points lower than that of DFLVX. Its Sharpe ratio was also lower, by about 10% in relative terms. The fund’s increased freedom to search the globe for undervalued stocks didn’t produce the desired outcome.
The next table shows the results of the most recent 15-year period, 2000 through 2014.
|
Fund
|
Annualized Return (%)
|
Standard Deviation (%)
|
Sharpe Ratio (%)
|
|
Tweedy Browne Value (TWEBX)
|
5.79
|
14.05
|
0.35
|
|
DFA U.S. Large Value (DFLVX)
|
8.44
|
20.55
|
0.42
|
As you can see, the return of TWEBX was 2.65 percentage points per year lower than that of DFLVX. And while it did experience lower volatility, DFLVX’s Sharpe ratio was 20% higher in relative terms.
Our final table shows the results for the full 21-year period from 1994 through 2014.
|
Fund
|
Annualized Return (%)
|
Standard Deviation (%)
|
Sharpe Ratio (%)
|
|
Tweedy Browne Value (TWEBX)
|
8.90
|
15.45
|
0.47
|
|
DFA U.S. Large Value (DFLVX)
|
10.44
|
19.10
|
0.50
|
DFLVX provided returns 1.54 percentage points higher than TWEBX and produced a higher Sharpe ratio. Also, in February 1995, DFA introduced U.S. Large Cap Value II fund (DFUVX), a lower-cost version of DFLVX, which is available to DFA clients who maintain a certain minimum level of assets with the firm. Currently, the expense ratio for DFUVX is 0.13 percent, 0.14 percentage points lower than the 0.27 percent expense ratio for DFLVX. Thus, investors with access to the lower-cost version of the fund would have had even better performance. That was true for each of the five-year periods I examined.
Summary
As I mentioned previously, Tweedy Browne Value has an outstanding reputation. And Morningstar shows that for the 15-year period ending January 6, 2015, it had a ranking in the 28th percentile of its peer group. In reality, the fund should look even better because Morningstar’s data doesn’t account for survivorship bias. Poorly performing funds either close or are merged out of existence, and their returns magically disappear. Thus, compared to most actively managed funds, Tweedy Browne Value is a winner.
But comparing Tweedy Browne Value to the average actively managed fund is false praise, because the average actively managed fund is a loser. TWEBX underperformed the well-constructed and passively managed DFA large-value fund. DFLVX’s ranking was in the 10th percentile. Even better, the lower-cost DFUVX had an 8th percentile ranking. It outperformed DFLVX by 0.14 percentage points a year due to its lower expenses.
Evidence like this demonstrates why Charles Ellis labeled active management a loser’s game in his 1998 book, Winning the Loser’s Game. It’s not that you cannot win, but the odds of doing so are so low that you are better off not even trying.
Even great investors such as Warren Buffett and Bob Goldfarb have a very difficult time identifying the future outperformers. If they have such a hard time, what are the odds you’ll succeed? What advantage do you have over them?
Larry Swedroe is director of research for the BAM Alliance, a community of more than 150 independent registered investment advisors throughout the country.
Read more articles by Larry Swedroe
