How Our Days Became Numbered
This is a carefully researched book by historian Dan Bouk, who examines how America came to terms with becoming simply numbers in a ledger. This change, however, took place voluntarily as a result of people attempting to reduce the risks of life through the purchase of insurance. The story begins after the Panic of 1873 and stretches through the Great Depression, although there is a concluding chapter briefly covering current issues. This review will provide a summary of the book, with greater emphasis on those chapters most relevant to forensic economists. However, it will be helpful to first briefly review some basics of life insurance.
Life insurance is a financial arrangement that transfers risk from the insured to a risk pool administered by an insurer. The risk pool should have several characteristics. First, the units should be relatively homogeneous. The idea that each person should be charged the same rate as others facing the same risk is called actuarial equity. Assume that in Company 1 both smokers and non-smokers are grouped together and charged the same rate. In a competitive insurance market, Company 2 may split this group into smokers and non-smokers. Non-smokers in Company 1 may realize they are being overcharged and therefore sign up with Company 2. The first company is suffering from what is known as adverse selection, given that some of the healthier members of Company 1 will leave while those remaining, on average, are less healthy. At this point, Company 1 may be forced to raise its rates. If it does not, or cannot due to government regulations, it could go bankrupt. The book identifies the process of placing individuals into appropriate risk categories as classing.
A second characteristic is that the risk pool must be diversifiable. A diversified risk is unique to pool members but can be reduced or eliminated by holding a portfolio of such members whose risks are uncorrelated. Floods and wars are generally not diversifiable risks because they may affect large numbers of the pool at the same time. A related point is that insurance is not provided unless the item being insured is truly an unpredictable risk. There is, in fact, no insurance against not successfully utilizing one's college degree because the person could put in less work effort than otherwise. This behavior is termed moral hazard.
Third, the pool must be large. By pooling more homogeneous units, you do not change the average loss but the standard deviation is lowered as determined by the law of large numbers. Bouk calls this process within risk categories smoothing the data. This is one of the major themes of the book–namely, the conflict between classing and smoothing. Insurance companies today deal with this tradeoff in a practical way. There must be sufficient classifications to properly take into account risk and avoid the problems of adverse selection. On the other hand, the classes must be of sufficient size to estimate future mortality rates so the law of large numbers can operate properly.
Chapter 1 focuses on classing, or assigning individuals to a risk class. The chapter begins with a discussion of the trial in 1878 of Thomas Lambert, whose company had failed financially. He utilized biometry, statistical tools measuring the body, to help predict mortality. Importantly, Lambert hoped to use biometry to predict the mortality of an individual, not a class of individuals. But perhaps the first function of insurance is risk pooling, which implies that there must be sufficient numbers in each risk class to allow for a reasonable prediction of mortality. What Lambert failed to recognize was that both classing and smoothing were necessary parts of the process.
This same tension arises when forensic economists calculate economic loss. The economist is focusing on a particular individual and usually requires specific information about that person. For example, if there is an injury case where the plaintiff has a well-documented history of earnings, a record of those earnings is the best source of information. However, in the case of a minor, there is no earnings history, and the economist will have to put the plaintiff into a larger “class,” usually related to an assumed level of education. The courts demand a focus on the individual, much as Lambert did in his insurance company, but sometimes that is not possible.
Chapter 2 is titled “Fatalizing,” which is the assumption that historical data are an accurate guide for predicting future mortality. Biometry (and phrenology) attempted to forecast future mortality for the individual while actuaries smoothed the data to predict mortality for a population's future. But both approaches utilized historical data for their predictions and hence both utilized fatalistic laws. This approach ran into difficulty once insurance companies began to expand into new areas of sales including women, children, and especially African Americans. The basic conflict was that the insurance companies saw the future of African Americans as a continuation from the Civil War while African American activists saw the War as a rupture where the future would be significantly different from the past.
At first, most insurance companies did not sell to African Americans. Metropolitan and Prudential insurance companies decided to sell to African Americans primarily through industrial insurance policies, but providing roughly two-thirds the benefits paid to the white population. The same policy was made with regard to women despite the fact that investigations showed that women lived longer than men. In 1884, legislation was passed in Massachusetts basically prohibiting any insurance company in the state from charging different rates based on race. Insurance companies fought the implementation of the bill, arguing that statistics bore out the economic necessity of risk-adjusting insurance premiums based upon race. When it was clear that defeating the bill would not work, almost all of the insurance companies in the North simply stopped selling insurance to African Americans.
Chapter 4 is titled “Smoothing” which, as discussed above, relates to averaging data within risk classes, including the estimation of dividends. The chapter begins by describing the New York City investigation of MONY (Mutual Life Insurance Company of New York), commonly known as the Armstrong Investigation. The chief investigator was Charles Evans Hughes, who later became governor of New York and Chief Justice of the Supreme Court.
The primary charge against MONY was that the company used money that should have been received by policyholders to line the pockets of the firm's employees. Further, this was due to the practice of smoothing where MONY averaged dividends over several periods rather than paying current dividends based on current performance. In fact, all insurance companies forecast many things that lend themselves to a smoothing process such as the future growth in paid premiums, the rate of withdrawals of existing customers, the load factor, the rate of return on investments, and many other unknown factors.
Given all of these unknowns and the state of technology in 1905, it was not surprising that the company could not adequately respond to all of the detailed questions posed by Hughes, such as the cost of maintaining the firm's investments. (After all, even today who can forecast future interest rates?) The insurance companies through smoothing and other techniques promised both protection from life's uncertainties as well as regular dividends despite the wild ups and down of the American economy. When that did not happen, the public wanted to know why. As a result of the investigation, legislation was passed that limited many of the practices of insurance companies.
Chapter 5 is titled “A Modern Conception of Death.” While Chapter 2 examined whether historical data could properly be used to forecast future events, this chapter looks at whether the future itself could be changed with a “correct” set of policies. Two leading individuals portrayed in the chapter were Louis Dublin, a biologist working at MONY, and Irving Fisher. Fisher is best known as one of the greatest American economists of the 20th century, but in the context of this book is portrayed as a health reformer.
Louis Dublin and Irving Fisher helped spearhead a new view of insurance. Fisher believed that certain practices of individuals and policies of government could significantly change the mortality rates of individuals. Some of them involved actions that people take on their own, such as better diet and getting exercise. Other suggestions related to curing certain diseases and improving sanitation. Fisher calculated what he termed “hypothetical life tables,” which estimated mortality rates if Fisher's recommendations were followed. Some of the suggestions could, in fact, be instituted easily by the insurance industry, such as requiring medical checkups or implementing fire prevention measures in housing.
In 1913, Fisher, along with Dublin and Lee Frankel, who at the time ran the welfare department at Metropolitan, started a new organization called the Life Extension Institute (LEI). Fisher hoped LEI would eventually be extended beyond Metropolitan since improving the health of the entire nation was one of his personal goals. For policyholders, it guaranteed periodic medical checkups. For the public, a book was published, modestly titled How to Live, mainly focusing on eating healthy, getting sufficient exercise, and having a positive attitude towards life. In the end, the insurance industry was not willing to accept this proactive approach. There was some good that came out of this. By the early 1920's, Metropolitan was beginning to sell health insurance.
Chapter 6 is titled “Valuing Lives in Four Movements.” The musical title relates to the chapter's discussion of Charles Ives, an early 20th century American composer. However, Ives was not in the chapter because of his musical talent but because of his position as head of the largest MONY insurance agency in New York City.
What the chapter describes is how insurance agents supervised by Ives went about convincing potential customers that they needed to buy life insurance and how much they needed. This meant telling a typical family what it would lose if the breadwinner unexpectedly passed away. For the forensic economist today, the following quote is frighteningly modern.
Risk makers traditionally fostered the ideal that each man should have enough life insurance to fully replace himself: enough to pay his survivors his current income minus the cost of maintaining himself (that cost being set in Ives's calculation at 36%). Translating that desire into life insurance took a bit of arithmetic and an understanding of how to determine the value today of dollars to be earned in the future. The agent, having completed the calculation, could tell a man the present value of all his future earnings minus his future expenses. This, the agent would argue, constituted each man's ultimate responsibility to his family, his paternal value. (Bouk, pp. 156-57.)
Bouk concludes the chapter by talking about the value of the statistics collected by the insurance industry. The statistics were certainly critical when selling their product and were even helpful to those who wanted to apply then in a larger social context. But ultimately it was recognized they had a limited value, especially in the latter context. Irving Fisher, the man instrumental in promoting LEI, had a telling comment about the Institute's medical director, Eugene Fisk, after he died at age 64. Fisher was asked why Fisk died prematurely. Fisher commented, “Life should be measured in terms of things accomplished, rather than in years.” (Bouk, p. 181)
Chapter 7 is titled “Failing the Future.” The eugenics movement reinforced discrimination that occurred in the insurance industry. Both Louis Dublin and Irving Fisher, who served as the first president of the American Eugenics Society in 1923, were strong supporters. Since the industry favored medical examinations and evaluations of individuals according to risk categories, why not extend that approach to a larger scale? For example, perhaps marriage licenses should not be granted unless the proposed couple was given a clean bill of health. Or immigrants should not be allowed into the country if they were mentally deficient. Such arguments were made before Congress; and in 1924, the Johnson-Reed Act limited immigration according to national origin.
The chapter concludes by examining the practices of insurance companies sold to African Americans. An investigator, Lawrence Brown, examined these specific practices. As noted earlier, most companies simply refused to sell policies to African Americans. Those that did generally charged a higher rate. The view of the insurance companies was basically that one should not discriminate unless there is an actual difference in mortality rates. The problem was that the differences were based on shaky data. For example, data for all white Americans were lumped together; and there were individual categories that had lower mortality rates than African Americans, such as the Irish. Further, as mentioned earlier, there was the problem of utilizing historical data to forecast the future.
The conclusion is titled “Numbering in Layers.” Insurance companies were finding new ways of categorizing risks. The Life Extension Institute, using newly refined data, attempted to improve the lives of citizens by recommending physician visits, losing weight, exercising, and other “healthy” practices. But it was the search for reliable wage and earnings data that proved the most challenging for many in the industry.
To accurately estimate the insurance needs of individuals, it was necessary to forecast their earnings. In addition, it was helpful to have data on unemployment rates and an estimated year of retirement. Dublin and Lotka attempted to put this information together, but there was little raw data. The actual collection of useful data fell to two former insurance employees now working for the federal government, Otto Richter and Bill Williamson. Unlike private pensions, Social Security required collection of data throughout a person's worklfe. Forensic economists can thank both the insurance industry, which started the process, and the federal government employees who made the record complete, for allowing a relatively full record of earnings for almost every plaintiff's case.
In summary, Bouk has written a powerful history of the insurance industry that should be of great interest to forensic economists. Of special note is the description concerning the value of man as emphasized by Dublin and Lotka. The model of forecasting the present value of future earnings minus personal consumption of the deceased is virtually identical to that used by forensic economists today. Of equal significance are larger questions that still remain unanswered. To what extent should past data be utilized in forecasting the future? What is the importance of “smoothing” the data as opposed to extracting data specific to the individual? How do you calculate human capital when it is not sold on a market, where the most notable example is household services? Another question that concerned Bouk throughout the book was how to account for different risks of groups such as African Americans when discrimination is one of the root causes for this result. The use of worklife tables and life expectancy tables based on race and gender has, in fact, been a subject of discussion among forensic economists.
In the very last sentence, Bouk leaves us with a message to ponder from Psalm 90:12, which reads: “So teach us to number our days, that we may present to Thee a heart of wisdom.” It is important to understand we have a finite time on earth and that each day is precious. Numbers are important, but ultimately we need to count our days so they can be filled with wisdom and meaning.