Household finance studies how households use financial instruments to achieve their objectives. Unique features of household finance problems include a long but finite planning horizon, nontraded assets, illiquid assets, borrowing constraints, and complex taxation. Many households invest efficiently, but a minority makes investment mistakes. Cross-subsidies in existing markets from naive to sophisticated investors may inhibit financial innovation.

Ideal household financial data would provide highly accurate information on individual households over time that is representative of the entire population, would present exhaustive details as well as total wealth, and could be disaggregated to individual assets. Necessary data are difficult to obtain primarily because individuals tend to guard their privacy and may have a complicated set of financial arrangements. Various incomplete datasets exist, but the absence of a complete set hampers household finance measurement.

Modeling household behavior is difficult because households face constraints not considered in standard finance texts. Examples of complexity are the following: (1) Idiosyncratic risk of nontradable human capital cannot be hedged; (2) housing, the dominant household asset class, is illiquid; and (3) households may face binding borrowing constraints in the present or future. Empirical models following Merton’s consumption and intertemporal capital asset pricing model explain some discrepancies between mean–variance analysis predictions and what financial planners advise. The assumptions of these models that assets are liquid and tradable, however, are contradicted by idiosyncratic unhedgeable labor income risk. The usefulness of these models is also limited by the presence of illiquid housing, borrowing constraints, and a complex tax structure.

U.S. data show that low-wealth households have little or no financial assets, and even those higher in the wealth scale have limited holdings of public equity. Housing is the dominant asset of the middle class. Equity is the dominant portfolio share only for high-wealth households that more willingly accept risk. Education, income, and wealth have strong effects on participation in public equity markets. The main influence on portfolio shares, both for public and private equity, is wealth. Nonparticipation in risky asset markets is an investment mistake that less-educated and poorer households are more likely to make.

Brokerage account data on U.S. asset allocations within each asset class show that many households own only a few individual stocks, but this risk may be offset by indirect ownership through mutual funds and retirement accounts. Many households own large holdings of their employers’ stock. There is a strong local bias in asset holdings.

Household data gathered by the Swedish government provide one of the most comprehensive datasets. These data show broadly consistent asset allocation patterns between the United States and Sweden. Sophisticated Swedish households tend to follow investment strategies closer to those recommended by standard financial theory. Overall, Swedish investors take substantial idiosyncratic risk, but the negative effect on the typical household’s welfare is modest. Like U.S. households, many Swedish households do not participate in risky asset markets. Household decisions to hold fixed-rate mortgages (FRMs) or adjustable-rate mortgages (ARMs) may result from borrowers’ perception that they lack the skills to invest efficiently. Households must consider variables such as the likelihood of moving, present and future borrowing constraints, real interest rate risk, and inflation risk within the context of the household’s risk aversion. Apparently, some households believe, incorrectly, that long-term interest rates are mean reverting, and they choose between ARMs and FRMs accordingly. Another investment mistake that households make regarding mortgages is the failure to refinance FRMs when conditions justify.

Slowness of innovation in retail financial markets is a puzzle. The author argues that existing products reward sophisticated households at the expense of unsophisticated households in a way such that no one is motivated to introduce simpler, new products. This cross-subsidy may inhibit financial innovation.

The author concludes that many households invest efficiently, but a minority of households that face complex investment decisions make serious investment mistakes. Serious mistakes are not participating in risky asset markets, underdiversifying risky portfolios, and not refinancing fixed-rate mortgages. Some markets involve cross-subsidies that may inhibit financial innovation.