One of two approaches is typically used when estimating the value of any company’s equity: fundamental valuation or relative valuation. Academic research and teaching tends to emphasize fundamental valuation models, which involve predicting future cash flows or future earnings and calculating their present value today. However, relative valuation models, which typically involve price multiples — ratios used to compare a company to a group of similar companies — are much more common in practice. A new study by Professor Doron Nissim compares the accuracy of different relative valuation methods in terms of their ability to explain the stock prices of insurance companies.
Outside the financial sector, the balance sheet and book value of equity are not viewed as a crucial basis — or as a particularly accurate basis — for valuation. But in the financial sector, valuation often starts from the balance sheet and focuses on the book value of equity. Nissim’s study shows that book value ratios perform relatively well when valuing insurance companies, and that over the last decade book-value ratios have performed significantly better than earnings ratios. These findings explain the focus of insurance companies, such as Warren Buffett’s Berkshire Hathaway, on book value and book value growth as the primary determinants of value and value creation, respectively.
While insurance companies are financial institutions, some unique features influence their valuation. Insurance companies tend to hold far more securities as a percentage of their balance sheet than do other financial firms — 70 versus 20 percent for banks, on average — and when insurance firms sell these securities, the treatment on the financial statements can influence valuation. Special capital regulation requirements that dictate how much capital and reserves insurance firms must keep on hand likewise can have significant effects on the balance sheet and the firm’s valuation.
Because of these and other unique features of the insurance industry, analysts often exclude Accumulated Other Comprehensive Income (AOCI) from book value (a practice unique to the insurance industry). AOCI measures unrecognized economic gains and losses that result primarily from changes in interest rates, credit spreads, or other factors that affect the value of the investment portfolio. Excluding these items is usually seen as a way to reduce the volatility of book value and accounting distortions. However, Nissim’s research shows that excluding AOCI tends to worsen, rather than improve, the accuracy of valuation. Changes in the value of the investment portfolio are only partially offset by changes in the value of insurance reserves (i.e., recognized obligations for expected claim payments) and therefore affect equity value and should be incorporated in the valuation. For example, an important factor contributing to the large declines in the stock prices of life insurance companies during the financial crisis was the increase in credit spreads. Credit-risky investments dropped in value, with little or no offsetting drop in the value of the insurance reserves. Most insurance companies excluded these losses from their earnings statements and the losses were reflected in AOCI. Because these losses were priced by investors, excluding AOCI from book value reduced the accuracy of book value–based price-ratio valuations.
Another surprising finding of Nissim’s study shows that excluding realized investment gains and losses from earnings does not improve valuation accuracy. An exception occurred during the recent financial crisis, most likely caused by an increase in gains trading or the selective realization of gains. Unlike other sources of insurers’ revenue — including premiums, investment income, and various fees — realized gains and losses are both discretionary and highly volatile. Accordingly, analysts and insurance companies often emphasize metrics such as operating income, which exclude realized gains and losses. However, to the extent that these items are used by insurance companies to smooth (or show steady rather than variable) reported income over time, they may actually increase the accuracy of reported earnings as a proxy for permanent income and so improve the precision of earnings-based price-ratio valuations. This appears to be the case in so-called normal times. In contrast, during the financial crisis some insurers selectively sold investments to artificially increase reported income.
Consistent with common industry practice, the study finds that considering return on equity when using the price-to-book ratio to value insurers significantly improves valuation accuracy. This result may help explain variation in price-to-book ratios over time and across companies. For example, the current valuation of Berkshire Hathaway, at a book-value ratio of 1.1, is significantly lower than the historical average of 1.6. The financial press has discussed alternative explanations for this apparent anomaly, ranging from market inefficiency to concerns regarding the company’s succession plan, price pressure from expected sales of donated shares, and other factors. “Perhaps it is an anomaly,” says Nissim. “Alternatively, it could be that investors project a decline in return on equity.”
Nissim’s findings underscore the need for investors to exercise caution in using price metrics in valuing insurance companies. “What is reflected in earnings and book value can change in ways that may change the price-earnings or price-to-book ratios,” he says. “Using Berkshire Hathaway as an example, if you expect profitability to be low compared to the past, the price-to-book ratio doesn’t have to be the size it was in the past. Investors should keep in mind that the price-to-book and price-to-earnings ratios are about future expectations.”
Doron Nissim is the Ernst & Young Professor of Accounting and Finance at Columbia Business School.