Difference in Calculated Betas using Monthly Data
The table below summarizes the Betas obtained from the regressions and the betas obtained from the Bridge (see Excel file for detail).
TABLE 1
BETAS for Du Pont compared with different “markets” / NYA / NYA / TMW / TMWComputed / Bridge / Computed / Bridge
Monthly
/ 1/95 - 12/97 / 1.098 / 1.153 / 0.9360 / 1.0581/98 - 12/00 / 1.014 / 1.113 / 0.5856 / 0.937
Daily / 29/12/97 - 29/12/00 / 0.849 / 0.759 / 0.4728 / 0.212
There are several factors determining a company’s Beta.
Operating Leverage refers to the proportion of the total costs of the firm that are fixed. Other things remaining equal, higher Operating Leverage results in greater earnings variability which in turn results in higher betas.
TABLE 2 (see Excel file for details)
Year / % in EBIT / % in Revenues / EBIT Variability Measure2000 / 5% / -14% / -290%
1999 / 8% / -4% / -48%
1998 / 3% / 3% / 132%
1997 / -44% / -62% / 139%
1996 / 4% / 3% / 61%
1995 / 7% / 14% / 191%
CHART 1 (see Excel file for details)
The higher degree of operating leverage, the greater the potential danger from forecasting risk. That is, relatively small errors in forecasting sales can be magnified into large errors in cash flow projections.
According to TABLE 1, the Beta calculated from monthly data in periods 1/95 - 12/97 and 1/98 - 12/00 differ slightly. In fact, the Beta in period from 1/98 to 12/00 is slightly less than the Beta derived from the 1/95 - 12/97 period. This is concurrent with the decrease in operating leverage (see CHART 1).
Good.
However, it is important to note that decrease in operating leverage is not proportional in decrease in Du Pont’s Beta. This implies that other factors have influenced the company’s beta offsetting a significant decline in the firm’s operating leverage.
TABLE 3 summarizes Du Pont’s annual Interest Expense beginning in 1995. CHART 2 is a graphical representation of the same data. As the chart indicates, the interest payments have been growing from 1997 steadily. This is a direct evidence of the increase of Financial Leverage for the period 1997-2000.
TABLE 3
Year / Interest Expense1995 / $ 758
1996 / 713
1997 / 389
1998 / 520
1999 / 535
2000 / 810
CHART 2
You should look at financial expense, standardized by something, such as EBIT. Even in levels, it is not clear that leverage is increasing. And what about just looking at the debt/equity ratio?
Changes in Financial Leverage have a positive relationship with the changes in a firm’s Beta. The more debt the company has the higher is the Interest Expense. Larger debt would imply the increase in the firm’s risk which would be represented by a larger Beta.
As such, since there was an apparent increase in Financial Leverage, the Du Pont’ Beta should have increased in periods from 1997 to 200, but it did not. Why?
The answer becomes apparent if we compare the changes in both Operating and Financial Leverage. While the Operating Leverage decreased, the Financial Leverage of the company increased. In effect, both changes have offset each other to certain degree causing Du Pont’s Beta to decline slightly.
We also have to remember that there were other factors that could have caused Du Pont’s Beta to change. One of them is the change in size. In a period beginning from 1997 to 2000, Du Pont has purchased several companies thus increasing its size and decreasing its beta. In addition, the nature of the business is also a critical element in determining the company’s beta.
According to Yahoo!, “Du Pont Company is engaged in science and technology in a range of disciplines including high-performance materials, specialty chemicals, pharmaceuticals, and biotechnology. The Company operates globally through some 20 strategic business units. Within the strategic business units, approximately 90 businesses manufacture and sell a wide range of products to many different markets, including the transportation, textile, construction, automotive, agricultural and hybrid seeds, nutrition and health, pharmaceuticals, packaging and electronics markets.” This indicates that the company is well diversified which also should bring the Beta down.
This is very good, but did DuPont become more diversified over time?
Calculated Betas: Monthly vs. Daily
As TABLE 1 indicates, the Beta calculated from daily data is much lower than the Betas calculated from the monthly data. In fact, the daily Beta is below 1.0, which implies that Du Pont’s stock is less risky than the NYSE and TMW if you trade them within a day. This is because the average daily return on Du Pont’s stock is positive while the average returns on the NYSE and TMW are actually negative (see Excel file).
This is not necessarily because the actual risk is lower. It has to do with non-synchronicity.
If we compare the beta of Dupont which was calculated by regressing Du Pont’s returns against NYSE returns with the Du Pont’s beta obtained with the regression against TMW’s returns, we will see that the latter beta is much lower. It is lower because average daily return for TMW is lower then the return on NYSE, which makes TMW a riskier investment, compared to returns of Du Pont. (The difference in returns is the result the composition. NYSE consists of about 3000 securities, while TMW consists of 5000 companies. This makes TMW a better representative of the market).
Actually, the composition of TMW and NYSE relative to DuPont is what would make the one beta or the other higher. It doesn’t necessarily mean that either one is automatically better. It depends on the use.
It is interesting to note, however, that the Betas calculated from monthly data are above one, suggesting that, over long term, the investment in Du Pont is riskier than a short-term investment. This argument is supported by the fact that average monthly return is slightly lower than returns on NYSE and TMW.
Bridge Betas: Monthly vs. Daily
According to the graphs obtained from the Bridge, Beta (daily) derived by the system using returns of Du Pont and TMW are much lower then the Beta (daily) obtained using returns of Du Pont and NYSE. In fact, this behavior mimics the behavior of daily Betas calculated by us using regression. Although the values obtained by us and values generated by the system are different, they represent the same pattern: Du Pont’s Beta at TMW as a market is almost a half of the Du Pont’s Beta at NYSE representing a market. (This support the argument before stating that, on a day-to-day basis, TMW is the riskiest investment compared with Du Pont and NYSE).
On the other hand, Monthly Bridge Betas for two different periods obtained by comparing returns on NYSE and TMW are almost identical (Beta for TMW during the period starting on 1/95 to 12/97 is equal to 0.989; while Beta for TMW during the period starting on 1/98 to 12/00 is equal to 1.00). Furthermore, both Monthly Betas indicate that NYSE and TMW indices behave similarly in terms of their returns on a monthly basis.
Bridge Betas vs. Calculated Betas
BETAS for Du Pont compared with different “markets” / NYA / NYA / TMW / TMWComputed / Bridge / Computed / Bridge
Monthly
/ 1/95 - 12/97 / 1.098 / 1.153 / 0.9360 / 1.0581/98 - 12/00 / 1.014 / 1.113 / 0.5856 / 0.937
Daily / 29/12/97 - 29/12/00 / 0.849 / 0.759 / 0.4728 / 0.212
The difference between Computed and Bridge data is direct result in different number of periods used to estimate the slopes on the scatter plots comparing Du Pont’s returns with those of NYSE and TMW.
How did you get a different number of periods? You’re looking at the same time period, aren’t you?
Although the values for Betas obtained by us through regression analysis and the values given to us by the system differ, both sets of Betas tell the same story. As TABLE 1 indicates (reproduced here again for you convenience), all Betas (Bridge and Calculated) obtained using daily data a below 1.00 which once again makes us to conclude that Du Pont’s stock is less risky than NYSE and TMW indices.
It is important to note that the Monthly betas for the period 1/95 – 12/97 are very similar and have a value very close to 1.00. However, it is apparent that behavior of Du Pont is different on a monthly basis. In fact, the data suggests that Du Pont stock becomes more risky than the indices.
Nevertheless, the picture is somewhat different during the period 1/98 – 12/00. During this period, Du Pont’s Computed and Bridge Betas compared to NYSE continue to imply the higher risk, while Computed and Bridge Betas compared to TMW indicate a lower risk for Du Pont than the risk in the market.
Your analysis is very good. You could, however, have used the regression statistics to make some of your points as well.
10 out of 10