Session 5: Betas (Relative Risk Measures)

Session 5: Betas (Relative Risk Measures)

Estimating Relative Risk

In this section, the speaker discusses how to estimate relative risk and introduces beta as a measure of relative risk. The speaker also mentions that there are alternative ways to measure relative risk.

Beta as a Measure of Relative Risk

  • Beta is a measure of relative risk that tries to determine whether a stock is more or less risky than an average-risk stock.
  • Conventional betas are derived from risk and return models fundamental to finance, but there are alternatives to conventional betas.
  • The three inputs for estimating the cost of equity for an average-risk company are the risk-free rate, equity risk premium, and beta.
  • There are different approaches to measuring relative risk, including beta, multiple betas, stock price-based approaches, and accounting measures.

Issues with Regression Betas

  • A regression beta is just one way of thinking about beta and can give different results depending on how the regression is set up.
  • The standard error of the beta indicates how wrong the coefficient could be. A typical standard error for a US company's beta estimate is 0.2.
  • A noisy estimate can overwhelm your beta estimate if you only use one regression or slice of data.
  • The worst beta estimates often come from the best-looking regressions.

Example: Nokia's Beta Estimate

  • An example shows Nokia's great-looking regression with high r-squared and low standard error but may not be accurate due to noise in the estimate.

Understanding Beta

In this section, the speaker explains how to think about betas in macro terms and what factors contribute to a company's beta.

Factors that Affect a Company's Beta

  • The type of business a company is in affects its beta. The more discretionary the product or service, the higher the beta.
  • The proportion of fixed costs in a company's cost structure also affects its beta. Companies with high fixed costs have higher betas.
  • Borrowing money creates fixed costs and magnifies risk, so companies with higher debt levels have higher betas.

Bottom-Up Betas

  • Instead of relying on regression betas, bottom-up betas are calculated by averaging betas across similar publicly traded companies and adjusting for leverage and business mix.
  • Bottom-up betas are more precise than individual regression betas because they average out mistakes.
  • Bottom-up betas can be used for private businesses and can capture risks that regression betas cannot.

Estimating Beta for SAP

In this section, the speaker explains how to estimate beta for a company using publicly traded software companies and consulting firms as examples.

Estimating Value of Businesses

  • The speaker condensed three businesses into two: software and consulting.
  • Publicly traded software companies and consulting firms were used to estimate unlevered beta for each business.

Additional Estimation Detail

  • Revenue weights can be used to estimate value, but it assumes that $1 in revenue in one business is worth the same as $1 in revenue in another business. This may not be true due to different margins.
  • The speaker looked up what multiple of revenues companies need to these businesses were trading at to get a better idea of how the market values them.

Weighted Average Unlevered Beta

  • The multiple of revenues for software companies was applied to the revenues that SAP gets from a software business, and the same was done for the consulting business. A weighted average was then calculated based on estimates that showed SAP is about 80% software and 20% consulting, giving an unlevered beta for SAP's businesses.

Final Step: Debt-to-Equity Ratio

  • The debt-to-equity ratio for SAP was looked up, which is only 1.4 percent since it does not use much debt. Using this ratio and the German corporate tax rate, an estimated beta for SAP as a company was calculated. This would be used instead of regression beta when valuing SAP.

Importance of Relative Risk Measure

  • A measure of relative risk is necessary when attaching discount rates to companies with different risk profiles; betas are just one way to measure relative risk.
Playlists: Valuation
Video description

Describes what a beta tries to measure and after critiquing the standard regression approach to beta estimation, I develop an approach for estimating betas for individual companies.