with a t-value of 0.90; the ranking procedure is not statistically significant in
the smaller, socially screened universe. One finds positive and statistically sig-
nificant ICs even using only a larger-capitalized, socially screened universe
when one applies the Beaton-Tukey estimation procedure.
- Vantage Global Advisors has been the adviser to a socially responsible fund,
the Lincoln Life Social Awareness Fund in its Multi Fund Variable Annuity
Family, which has produced a net return of 16.40 percent for the seven years
ending March 13, 1996, whereas its socially responsible benchmark, the S&P
500 less its restrictions, has generated a corresponding return of 14.62 percent,
respectively. Vantage has used a quantitative proprietary model emphasizing
growth at a reasonable price (GARP) and will not invest in securities of firms
that (1) engage in activities that damage the natural environment; (2) produce,
design, or manufacture nuclear power or equipment for the production of nu-
clear power; or (3) manufacture or contract for military weapons; or (4) are in
liquor, tobacco, and gambling industries. It is indeed possible to be a socially
responsible manager and outperform the market. The seven-year returns are
annualized. The performance figures include the reinvestment of dividends and
other income. Past performance is not indicative of future results.
- If one believes that BARRA value and growth factor returns can be forecast for
the coming quarter using a Box-Jenkins (1976) time series model, then a random
walk with drift formulation with a seasonal moving average operator can in-
crease the CIBF (Consensus IBES Forecasted) weight when the BARRA growth
factor return is expected to rise relative to the BARRA value factor return and can
increase the predictive power of the model from a monthly IC of 0.052 (t-value of
1.66) to 0.063 (t-value of 1.99) during the 1987–1994 period.
- If one regresses 12-month total returns for the 1992–2002 period for all Rus-
sell 3000 stocks, one finds the following “social screening costs” (i.e., positive
returns):
Social Screen Cost T-Stat
Alcohol –.017 (–.40)
Military .036 (2.67)
Nuclear Power –.054 (–2.86)
Gaming, Tobacco .028 (0.77)
- If one uses net KLD social strength, the strengths less the concerns, and re-
gresses 12-month returns for the Russell 3000 stocks with the KLD criteria for
the 1992–2002 period, one finds that the net product strength variable, a sur-
rogate for R&D activity, is positively associated with total returns. Diversity
and employment net social strength measures create stockholder wealth losses.
Net Social Strength Coeff T-Stat
Community .021 (3.28)
Diversity –.008 (–2.12)
Employment –.021 (–4.40)
Environment .011 (2.31)
Product .020 (3.35)
270 NOTES
