Skip to content
All library documents

Why Debt-Financed Equity Trades Are Not Risk-Free Arbitrage

Article Quant Q&A · Author: Borun Chowdhury

Summary

The document examines an apparent arbitrage: buy an unlevered firm, have it issue debt, then sell the revalued equity. Under a simplified tax model, debt creates a tax shield that can increase total firm value and produce cash after the transaction. The answer explains why this result depends on restrictive assumptions rather than establishing a risk-free profit.

Potential offsets include bankruptcy costs, agency conflicts such as debt overhang that can reduce operating cash flows, and other ways firms may reduce taxes. The discussion places this tax-benefit-versus-distress-cost framework in the early capital-structure literature, while noting that later theories offer competing explanations for firms’ financing choices. It cautions that capital structure is not a settled problem: many theories are hard to test, and empirical models based on them have limited predictive power. The document gives a conceptual explanation, not a quantified trading strategy or evidence that a specific leverage trade is profitable.

Key ideas

  • Debt tax shields can raise firm value when interest is deductible.
  • The apparent arbitrage relies on assumptions that exclude costs and changes in cash flows.
  • Bankruptcy costs and debt-related agency problems can offset tax benefits.
  • Alternative theories of capital structure remain difficult to test and predict.

Tags

Full text
# Buy firm, leverage and sell, seems like arbitrage strategy. What's wrong with this argument?


# Buy firm, leverage and sell, seems like arbitrage strategy. What's wrong with this argument?












I recently came across with someone valuing a firm with different D/E ratios. My question is that this looks a bit spurious in that its mechanical. For instance if one wants to buy an unlevered firm and finds a value $V_U$ for its equity (there is not debt at this point). If they take debt then the firm value changes to $V_L=V_U + Dt$ and in particular equity value changes to $V_U- D(1-t)$. Then they sell the equity. The cash flow is

$$ -V_U + D + (V_U- D(1-t))= Dt $$

where the first term is the money spent to buy unlevered equity, second is the cash gotten by taking debt and the third is the money obtained by selling off the equity.

This looks like arbitrage. What's wrong in this strategy though?

## Answer by Matthew Gunn (score 3, accepted)

https://quant.stackexchange.com/a/38010

This sounds like a classic, introductory corporate finance question.

If interest payments on debt are tax deductible, increasing debt lowers corporate income taxes. Taking total firm cash flows as given, increasing debt effectively redirects cash flows from the government to equity holders. With no counteracting forces, firm owners will choose an all debt firm.

Why not do this? There are numerous objections you can raise to assumptions behind the above argument. For example:

- The cost of bankruptcy isn't zero.

- Excessive debt can cause agency problems (eg. underinvestment due to debt overhang) and reduce firm cash flows.

- There are numerous other ways to avoid taxes.

The list goes on... Anyway, these issues will be discussed in any corporate finance textbook or intro corporate finance class.

#### Some broader context...

The early academic, capital structure literature tried to explain optimal firm leverage as a tradeoff between tax benefits and bankruptcy costs. A lot though has happened since the early 1970s. There is now a tremendous quantity of corporate finance theory that posit alternative explanations for firm financing decisions.

Rather than give some simplistic survey of that theory here, I'd recommend the reader start with any introductory corporate finance text or class. The only thing else I'd mention is that the theory of firm capital structure is not a solved problem. Many existing theories are difficult to test, and to the extent that we can build empirical models to predict leverage ratios from theory, the $R^2$ you get aren't overwhelming.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.