Higher leverage leads to higher ROE, which (assuming no shenanigans), ought to lead to higher FCFE. However, risk goes up too, so P/CF multiples should be about the same at low levels of leverage, because additional return is balanced by additional volatility, but P/CF multiples ought to drop with higher leverage to reflect increased risks of not meeting interest payments if the debt levels get too high.
and one more thing bchad, because i love to read you type, is this statement true in all scenarios? "Adding leverage to the capital structure will make the stock more volatile because all of the cash flow or net income is now going to be measured off of a smaller equity value. " assuming the company has the same growth opportunities before and after debt issuance (i.e. no incremental upside) and assuming the debt does not put the company at risk (i.e. no incremental downside), wouldn’t the % change in the levered and unlevered company relative to the market remain the same? i understand that cf/earnings are going to a smaller portion of equity, but the stock should increase relative to the increase in debt (i.e. d/e = 1, stock should double overnight)
It seems that securities prices of leveraged vs. non-leveraged firms should be more volatile. I think that sophisticated investors would load up on more levered firms when outlooks are positive bidding up their price relative to the benchmark. When outlook is negative, they would sell them as their risk increases relative to the benchmark. So relative to the index, the more levered firms by definition magnify the earnings to equity wether to the upside or downside. Again, its only what I think but if one calculates betas over at least a complete economic cycle, leverage should be reflected in beta. That said, I dont think Leverage is the driving force of beta.
Matt, I’m not sure what you mean by "wouldn’t the % change in the levered and unlevered company relative to the market remain the same? " Take a look at the volatility of earnings or cash flows. Say that you have a $100MM company that is 100% equity and the quarterly ROA growth is 1% with a standard deviation of 2%. This means that the 95% confidence interval for NOPAT has a width of +/- $4MM (=2*2%*$100MM) Since the company is 100% equity, ROE = ROA = 1% Now let’s take the same company, but assume that the CFO targeted a capital structure of 50% debt and 50% equity. The company is still $100MM, but now $50MM is debt, and $50MM is equity. For simplicity, let’s assume that interest payments are zero; in reality, they won’t be, but it doesn’t change the main message here, unless the interest payments are so high that they consume all earnings. ROA is still the same, because ROA uses pre-interest earnings and is independent of leverage. ROA=1%, and expected earnings are $1MM, as before. The variability of NOPAT is still +/- $4MM, because how levered the company is shouldn’t influence whether they are able to sell their product and what their operating profit margins are. But now, let’s compare the volatility of earnings vs. equity values for the unlevered company to the levered company. 100% Equity : (+/- 4MM) / ($100MM Equity) = +/- 4% ==> 2% volatility (expressed as a SD) 50/50 debt/equity : (+/- 4MM) / ($50MM Equity) = +/- 8% ==> 4% volatility (as SD) Meanwhile ROE (100% equity) = $1MM / $100MM Equity = 1% ROE (50% Debt; 50% equity) = $1MM / $50MM = 2% Result: Unlevered company has half the ROE and half the Volatility of the 50/50 levered Levered company has twice the ROE and twice the volatility of the unlevered. In the absense of interest payment considerations, these companies are equally attractive on a risk-adjusted return. This is all about ROE, so how does this pertain to actual market returns? If the company pays all earnings as dividends, then the return is ROE/PB, where PB is equal to the price-book ratio. If the company retains all earnings, then the return is just equal to ROE. If there is a payout of dividends, then you get a weighted average of the two. In either case, the effect of leverage is the same - volatility of returns rises proportionally. All of this assumes that the PB ratio stays constant over the holding period. So the interesting question is what makes the PB value of a company change from period to period? In reality, the ROE will be eroded by interest payments, because NOPAT margin doesn’t include interest, while Net Margin does. Volatility won’t necessarily be affected by interest payments, except to the extent that outside investors decide that high debt is extra unattractive. Since interest payments add liquidity risk to the company and erode earnings (risk of insufficient cash to make interest payments) the risk-adjusted return of the levered company will be less than the unlevered company. You can probably recapture some of that lost risk-adjusted return by owning some of the company’s debt along with its equity. Taking on debt can lower a company’s WACC, which raises the total value of the firm’s future earnings. The PV of future cash flows is not captured very well by an analysis of the balance sheet, and so this often justifies the use of debt to maximize the market value of a stock. From a balance sheet perspective, there is no real advantage to levering up in terms of risk-adjusted returns. However, there may be economies of scales, and competitive and market share advantages to being a larger company that are not captured by balance sheet analysis.
OP: Beta is undeniably influenced by leveraged. In simple terms, Beta (j) = standard deviation (j) * correlation (j, M) / standard deviation (m). Now think of it intuitively: standard deviation (j) is a measurement of a stocks volatility. What kind of stock is more volatile: one with a 99% debt ratio or one with no debt?
in my opinion, there are 2 types of people who often criticize passionately the concept/derivation of beta, CAPM, the Black-Scholes model, etc., etc. - academicians with deep understanding of the subject, and amateurs with only basic understanding of the models. it’s so easy to point out the obvious “fallacies” or unreasonable assumptions in these models - “hey, the constant volatility assumption in Black-Scholes is $hit, volatility is not constant at all - here, i graphed it!”, or “CAPM is crap - it predicts every rational investor should buy the market portfolio. here, i just bought Google stock and i feel pretty good about it. Take that, CAPM!” it’s so much harder to appreciate how these relatively simplistic models, with all of their “unreasonable” assumptions, still work remarkably well - and they do!
bchad: beta measures changes in the stock’s market cap, not its bs equity. if the company’s market cap is $100 MM and we replace 1/2 the shares with debt, the market cap remains at $100 MM, but the shares are worth twice as much, as the attributable cashflow/earnings for the equity portion is the same (assuming no interest costs). its even possible that the equity would be worth more than double and its market cap would increase. if apple’s outlook were to remain constant for both scenarios, wouldn’t apple’s levered market cap adjust to the point where any additional expected growth on cash flow to equity is reflected in the market cap? thus, you would have to rely on changes to an outlook before an unlevered and levered apple have differing betas? i think sflcfa hit it on the nose that leverage will only magnify fp/es and fp/cf multiples of debt-laden companies in up markets and dwarf those ratios in down markets. if earnings were expected to stay constant, beta would be unaffected as the market cap would adjust to the point where changes in market cap would be the same for both the levered and unlevered versions of the company. it is the combination of a expected growth beyond what is realistic on a greater amount of cash flow / earnings to equity in boom times AND expected decline greater than what is realistic in bear times that will cause the levered beta to be higher. these are behavourial factors.
"bchad: beta measures changes in the stock’s market cap, not its bs equity. " Yes, Matt, I know, which is why I added that this assumes that P/B stays constant. I don’t often work with PB ratios, so I don’t know how volatile they are over time. It does seem to make sense that PB ratios shouldn’t fly all over the place, though.
sorry bchad. i didn’t mean to say “you’re an idiot, na na na nana na”, i just like to argue. hehe. thats why i’m in this business 
my argument proves that until there is news which affects the company’s profitability/growth, an increase in debt does not affect beta. your argument proves that in the long-term, where new information will eventually hit the airwaves, leverage should exaggerate changes in expected growth and earnings, and thus have a higher beta. its nice to get back to the old af where cfa-relevant topics are discussed.