Insights / Research BriefJan 17, 2024

Asset Demand of U.S. Households

Xavier Gabaix, Ralph S. J. Koijen, Federico Mainardi, Sangmin Oh, Motohiro Yogo
New data on households’ portfolios, including ultra-high-net-worth (UHNW) households, indicate that households play a limited role in stabilizing fluctuations in financial markets.

Recent research has revealed that in real-world markets, changes in asset prices do not lead to as large changes in the quantity of these assets bought or sold as standard theories have predicted. To date, this work has used data on institutional holdings and trading, however, and it is an open question whether households, and in particular very wealthy ones, play an important role in stabilizing fluctuations in financial markets. In this paper, the authors study this question using new data on the trading behavior of households. 

The authors use data from Addepar, a wealth management platform that specializes in data aggregation, analytics, and reporting for households’ portfolios. They collect monthly security-level data on portfolio holdings, flows, and returns of US households for the period from January 2016 to March 2023. Their data feature broad coverage across the wealth distribution – including ultra-high-net-worth (UHNW) households – and span multiple asset classes, covering both public and private assets. This allows the authors to study the flows and rebalancing behavior across asset classes and individual assets to determine whether households play an important stabilizing role in financial markets. They find the following:

  • Less affluent households tend to sell U.S. equities amid market downturns, while UHNW households tend to buy. This behavior is more pronounced among households who rebalance their portfolios more frequently.
  • Despite these patterns, the aggregate household sector plays a limited role in absorbing financial fluctuations. This is because the sensitivity of flows to returns is generally quite small, and the trades of different wealth groups partly offset each other. A 10% increase in market returns is associated with a change in flows of about 0.1% in absolute value.
  • A household’s flows to U.S. equities are negatively correlated with its “active returns” (the difference between an investor’s return and the market return). In other words, when a household’s portfolio is underperforming the market, the household’s flows to U.S. equities increase, and vice versa. However, the flows to U.S. equities of less affluent households are also positively correlated with broad market returns, meaning that these households tend to invest more in equities when the overall market is performing well and invest less when the market is down.  
  • Across all asset classes, three factors explain 81% of all variation in portfolio rebalancing. These factors take advantage of broad factors: long-term equity premium, or the extra return investors expect to earn by investing in the stock market over the long term; credit premium, or the additional return expected for investing in corporate bonds, which carry a higher risk of default than government securities; and lastly, municipal bond premium, or the extra yield that investors require for holding municipal bonds, which is influenced by factors such as tax benefits and the creditworthiness of the issuing municipality.

The facts documented here paint the picture of quite inert households (even for the extremely wealthy households), with low turnover and reaction to the aggregate stocks market developments, consistent with models of inertia, inattention and inelasticity. They provide important quantitative inputs into the design of structural macro-finance models with heterogeneous households and multiple risky asset classes and should assist in the writing of such models in future research.