TL;DR
This paper reveals a significant disconnect between how much life insurance older households have and how much they need to protect against financial hardship if a spouse dies. While, on average, insurance seems sufficient, a closer look shows many with high vulnerabilities are underinsured, while others with low risk are overinsured, leading to widespread, unaddressed financial risks.
What Problem Does This Paper Address?
The authors investigate a critical question in personal finance: are households adequately insured against the financial shock of a spouse's death, particularly as they approach retirement? The prevailing wisdom in the insurance industry is that life insurance is "sold, not bought," suggesting that purchasing decisions might not align with actual financial needs. This could stem from inertia, psychological factors, or simply a lack of understanding of one's own financial vulnerabilities. Previous studies have shown that a significant portion of surviving spouses experience substantial declines in living standards, but this paper aims to quantify the extent to which life insurance holdings mismatch these underlying vulnerabilities. They are particularly interested in understanding why this mismatch occurs and what household characteristics are associated with it.
Methods
The core of the paper's methodology involves constructing an elaborate life-cycle planning model. This model simulates household finances over time, accounting for a wide array of demographic, economic, and financial factors, including:
- Survival Contingencies: The model explicitly models different scenarios based on whether the husband, wife, or both survive.
- Income Streams: It uses actual Social Security earnings histories and a detailed benefits calculator to estimate survival-contingent income.
- Consumption Smoothing: The model aims to determine the "highest sustainable living standard" for each household, which is the level of consumption that can be maintained across all possible survival states without depleting resources.
- Vulnerability Measurement: Financial vulnerability is quantified as the percentage decline in this highest sustainable living standard that would occur if a spouse died. This is termed "IMPACT."
- Benchmark Insurance: A "benchmark" level of life insurance is calculated as the amount needed to ensure no decline in living standards for survivors.
- Actual vs. Benchmark: The study then compares actual life insurance holdings to this benchmark and to the calculated vulnerabilities.
The authors use a sophisticated financial planning software, ESPlanner, to implement this model.
Datasets
The primary dataset used in this study is the 1992 wave of the Health and Retirement Survey (HRS). This survey provides rich information on a wide range of household characteristics, including health status, income, wealth, pensions, Social Security benefits, demographics, education, housing, consumption, family structure, employment, retirement plans, cognition, and life insurance. The authors also utilized a matched, supplemental dataset containing Social Security earnings histories.
Key Insights and Ideas
- The "Average" is Misleading: While the average life insurance coverage might appear adequate to cover potential declines in living standards, this masks a significant underlying problem.
- A Startling Mismatch: There's a pronounced disconnect between who needs insurance and who has it. Many households with high financial vulnerabilities have surprisingly little insurance, while those with low vulnerabilities often have more than they need.
- Underinsurance is Widespread: Despite the average, a substantial portion of households face significant or severe financial consequences upon a spouse's death due to insufficient insurance.
- Vulnerability Varies Systematically: The degree of financial vulnerability and the propensity to purchase life insurance are not random; they vary systematically with individual and household characteristics.
- Insurance Doesn't Fully Offset Risk: On average, life insurance only offsets about 10% of the variation in financial vulnerability. For those most at risk, insurance coverage is often far from sufficient to maintain their pre-death living standard.
Key Results
- Average Decline: Without insurance, the average decline in living standards for surviving spouses was:
- Wives: 17.6%
- Husbands: 6.6%
- Secondary Earners: 20.3%
- Primary Earners: 3.9%
- Impact of Actual Insurance: After accounting for actual life insurance holdings, these average declines were significantly reduced:
- Wives: 4.49%
- Husbands: 1.81%
- Secondary Earners: 7.00%
- Primary Earners: 0.70%
- Underinsured at Risk: For those facing severe financial consequences (a decline of 40% or more), only about 25-30% held sufficient life insurance to avert these consequences.
- Poverty Rates: Taking insurance into account, poverty rates for surviving wives and husbands would be 22.5% and 14.6%, respectively. Without insurance, these rates would be 27.6% and 16.0%.
- Correlation Weak: Insurance holdings showed a weak correlation with financial vulnerabilities. In some regressions, those with lower vulnerabilities were more likely to have insurance.
Limitations and Future Work
- Model Abstraction: The life-cycle model, while elaborate, is still an abstraction and relies on imperfect information about household circumstances.
- Omitted Factors: The analysis omits uncertainty about future income and non-discretionary expenses (like medical care), which could lead to an understatement of insurance needs. The possibility of remarriage after a spouse's death is also not modeled, though it's considered less relevant for the age group studied.
- Data Noise: Measures of household assets can be noisy, potentially exaggerating the tails of the IMPACT distribution.
- Future Work: The authors suggest that further research could explore the role of remarriage and the impact of other economic variables measured with error. They also note that the findings are sensitive to assumptions about household economies of scale, pension survivor benefits, and housing flexibility.
Summary updated