Economics 13/02/2010: On benefits of marriage to investment

Feeling marital around the Valentine's Day? Well, how about investing in some stocks?

A paper just published in The Review of Financial Studies (2010, 23(1) pages 385-432) titled: “The Effect of Marital Status and Children on Savings and Portfolio Choice" by David A. Love (not kidding there) looked at the impact of marital status on optimal decisions about saving, life insurance, and asset allocation. It turns out, quite predictably I must add, that changes in marital-status and the number of children can have “important effects on optimal household decisions”:

• Widowhood induces sharp reductions in the portfolio shares in stock, and the impact is largest for women and individuals with children.
• Divorce causes men and women to reallocate their portfolios in different directions; men choose much riskier allocations, while women opt for safer ones.
• Children play a fundamental role in the optimal portfolio decisions. Men with children, for example, increase their shares in response to divorce by less than half as much as men without children.

Obviously, this might have something to do with the fall-off in disposable income and wealth, induced by a divorce when long-term child payments are involved. “…Imagine that a couple with two children living at home gets a divorce. Depending on the legal assignment of custody, the mandated level of child support, anticipated earnings, and the division of household wealth, the ratio of wealth to the present value of future income for each spouse will almost surely differ from the level previously observed in marriage. …this ratio is key to understanding optimal asset allocation because it summarizes, at least in part, the exposure of future consumption possibilities to fluctuations in financial markets.”

In addition to wealth-to-income ratio, divorce and portfolio choice are linked through changes in financial background risk as “the former spouses move from living on a combined income to each relying on a potentially more volatile single stream. …Uninsurable background risk, arising, for example, from labor income, business income, and housing, can have a quantitatively large impact on optimal portfolio decisions.”

A final way that the divorce might influence optimal portfolio choice is through its effect on savings “as the former spouses update their desired consumption of housing, food, transportation, and childcare.” Divorcees from a single car household buy a new car. They also increase childcare expenditure in most cases, unless large divorce settlements induce one parent exit from the labour force. Food consumption expenditure and volume rise, as all other consumption of both durables and non-durables.

Contrast the economic implications of divorce for the two-child couple with those of a childless couple. Members of the childless couple will still experience a change in wealth and income in the event of a divorce, “but there will be no additional shock to resources due to child support, college expenses, or differences in scale economies related to the assignment of custody. Given this differential impact on resources, it is reasonable to expect that the childless couple will respond differently to divorce in terms of saving and portfolio choice. In addition, children may also alter households' responses to widowhood. For example, depending on the strength of the bequest motive, surviving spouses with children will tend to have larger amounts of wealth relative to income compared to those without children, leaving them more exposed to market risk.”

All of these conjectures are supported by evidence, but there are some surprises in findings as well: “We find that households with children tend to accumulate substantially less wealth during the working years but that their slower rates of drawdown in retirement leave them with more savings toward the tail end of life.” Does marriage really mean life-long prudence?

“This trajectory of wealth accumulation is mirrored, in part, by the evolution of portfolio shares. Earlier in the life cycle, households with children hold riskier portfolio shares (by about 10 percentage points) than households without children, but the relationship reverses in retirement.” So no, risk aversion is lower for married couples probably because their dual incomes act as hedges against single income volatility.

Instead – it is bequest motive that drives them to become more conservative in older age. “…a riskier allocation for these younger households is optimal because their consumption streams are less dependent on the performance of financial markets. In retirement, however, children provide an incentive to maintain wealth for bequests, and the resulting increase in the wealth-to-income ratio makes households increasingly sensitive to stock market volatility.”

Hmmm, this brings us to taxes, then. A rise in inheritance tax during the wealth accumulation period of household life cycle implies a reduced incentive to save for bequest. This, should then result in lower risk aversion for older age households. And that, in turn, will lead to greater volatility of investment and also to higher cost of borrowing by the sovereigns. How so? Because if older households become less risk averse, their share of government bonds in total investment portfolio will drop. This means lower demand for bonds and higher yields on new issuance. Cost of sovereign borrowing goes up and the benefits of higher taxes to Government revenues are cancelled out, at least in part.

Imagine that – some say there is no such thing as Laffer Curve… not even at 100% marginal tax rate?