Despite strong headwinds, including renewed divestment pressure,1 the tobacco industry has proved quite resilient financially and outperformed the stock market over the past 18-1/2 years. So much so, that some institutional investors are now thinking of lifting tobacco bans in their investment policies. We found that most of the gains associated with holding tobacco stocks over this period were not specific to the tobacco industry, and could have been obtained in other ways. We also show it would have been possible to divest from tobacco without taking a hit to portfolio performance during our sample period.
WHY “SIN” HAS BEEN “IN”
Recent studies of “sin stocks” have shown their performance can be largely explained by standard asset pricing factors, such as profitability and investment.2 From December 1999 through July 2018, tobacco industry stocks in the MSCI ACWI Index outperformed the index as a whole by 11.7% annually.3 Using the MSCI Global Total Market Equity Model to disentangle the sources of this outperformance, it turns out it was mostly due to style factors (5.5%) and stock-specific sources (4.7%).
Performance of ACWI Tobacco industry simulated portfolio
Data from Dec. 31, 1999 to July 31, 2018
Specifically, tobacco stocks persistently benefitted from having lower beta, higher dividend and earnings yield, as well as better profitability and investment quality – in line with the observations by Blitz and Fabozzi. The results showed that the contribution from industry factors was less important. That is partly due to the fact that the model does not have a pure tobacco industry factor, so some of the movement of tobacco stocks is captured by stock-specific return.
WITHOUT TOBACCO STOCKS, DID PERFORMANCE GO UP IN SMOKE?
What may be more relevant to asset owners is the flip side of this story: Excluding the tobacco industry led to minor losses compared to the benchmark. The ACWI ex Tobacco industry portfolio underperformed the benchmark by 11 basis points (bps) annually over our 18-1/2 year study period. As a mirror image of the exhibit above, albeit at a smaller scale, most of this loss was due to style factors (-5 bps) and stock-specific returns (-3 bps).4 The total loss was fairly small, given the low weight of the tobacco industry in global benchmarks, but considering the level of tracking error (19 bps) and the nearly 19-year-long time period, it was statistically significant.
Performance of ACWI ex Tobacco industry simulated portfolio
Data from Dec. 31, 1999 to July 31, 2018
OPTIMIZATION HELPED EXCLUDE TOBACCO STOCKS WITHOUT LOSING THEIR BENEFITS
Quantitative managers may be interested in creating diversified portfolios that have no tobacco stocks but keep the same style exposure profile as their parent index with minimal tracking error. In other words: a style-hedged ex-tobacco portfolio. We present here one such example. Staying conscious of transaction costs (maintaining a stable style profile leads to more turnover), we estimated that increasing turnover by more than 10% would cost more than the losses we sought to eliminate.5 In our backtest, we set the turnover limit for quarterly rebalancing accordingly to 3%.6
The resulting backtested portfolio eliminated the loss due to style factors (see exhibit below). In this example, the portfolio also outperformed the benchmark by 43 bps annually, of which 39 bps was due to stock-specific sources. This should not be expected to happen in all settings.7
Performance of style-hedged ACWI ex Tobacco industry portfolio
Data from Dec. 31, 1999 to July 31, 2018
While turnover costs would have had to be managed closely, with the use of an optimizer, institutional investors could have created tobacco-free portfolios during the sample period that closely matched the style profile of their benchmark with minimal tracking error.
1 See TobaccoFreePortfolios
2 Blitz, D. and F. Fabozzi. (2017). “Sin Stocks Revisited: Resolving the Sin Stock Anomaly.” Journal of Portfolio Management, Vol. 44, No. 1, pp. 105-111.
3 Tobacco industry representation in the MSCI ACWI Index consisted of 20-30 stocks and between 1% and 2% of market capitalization during the sample period.
4 Average active style exposures ranged between -0.01 (dividend yield and profitability) and 0.01 (beta).
5 Based on a conservative estimate of 25 bps per percent of two-way turnover
6 The MSCI ACWI Index had roughly 2% annualized one-way turnover.
7 This report contains analysis of historical data, which includes hypothetical, backtested or simulated performance results. There are frequently material differences between backtested or simulated performance results and actual results subsequently achieved by any investment strategy. Past performance — whether actual, backtested or simulated — is no indication or guarantee of future performance. None of the information or analysis herein is intended to constitute investment advice or a recommendation to make (or refrain from making) any kind of investment decision or asset allocation and should not be relied on as such.
Further Reading
Investing for the long run: ESG and performance drivers