There’s No Place Like Home

With the invisible but inescapable global reach of the Internet, worlds of information are a keystroke away. But not all information is created equal. Whoever first said “Too much information, too little wisdom” described the key challenge of our time. Even if all information is received at the same rate and intensity, some elements will be more attractive than others for a given user. The reasons for this can be as simple as the message appearing in the user’s own language, popping up in a commercial on a popular TV station, or encompassing a full-page ad in a favorite magazine.

With higher value placed on more accessible information, the balance is changed and a bias emerges. What may seem to be small differences can add up to an important investment result—money stays at home. This phenomenon—called “home bias”—is well known in financial analysis: investors strongly tilt their portfolios toward domestic assets and away from foreign instruments.

“When managers differ in their ability to generate substantial returns, and investors do not know this ability but learn about it by observing past performance, investors will learn faster about the domestic funds,” says Wioletta Dziuda (Assistant Professor of Managerial Economics and Decision Sciences at the Kellogg School of Management). “Therefore, the domestic funds will be less risky for the investors. As a result, investors will channel more money to the domestic funds.”

Dziuda and her colleague Jordi Mondria (Assistant Professor of Economics at the University of Toronto) have found that portfolio managers can carry this home bias in investment up to the level of mutual funds, based on small amounts of market performance information distributed asymmetrically among individual investors.

Domestic Information Is Cheap Information
There is a simple explanation for home bias in individual investment—investors have better information about their domestic economy than about foreign ones. But it is puzzling that home bias persists even when the investment is undertaken by mutual funds, which are capable of accessing and analyzing information from virtually any source with equal effectiveness.

“We agree that almost all information is available to an investor who has time and resources to get it,” says Dziuda. “The problem is that some information is much ‘cheaper.’ While reading domestic newspapers, watching domestic TV, and talking to friends, individual investors easily receive information about how the domestic economy is doing. They don’t have to search.

“On the other hand,” she continues, “following the news of a foreign country requires time and resources that individual investors do not always have. The mutual funds, however, should have enough resources, and therefore we assume that mutual funds do not face an asymmetric information problem.”

Two principals are involved in mutual funds’ investment processes: managers, who run the funds, and individual investors, who hand over their hard-earned money for what they hope will be rewarding returns. And while we might assume intuitively that investors should distribute their wealth equally across domestic and foreign mutual funds, Dziuda and Mondria find that home bias persists.

Reinforcing Home Bias
Dziuda and Mondria built a mathematical model for home bias that describes a full circle. It begins with a home bias outlook in the individual investor’s access to information, follows its effect on investors’ judgment of a portfolio manager’s capabilities, and comes all the way around to reinforcing the home bias phenomenon itself.

They modeled a two-period economy, involving two hypothetical countries—one domestic, one foreign—and a continuum of investors. The domestic market represents any market one chooses to analyze. The foreign market, by contrast, represents any or all markets that are unfamiliar to domestic investors. For individual investors, the primary goal is to allocate their investments in the best possible way. Thus, individual investors would like to understand the ability of mutual funds to generate abnormally good returns.
 
Dziuda and Mondria began with two main assumptions, believing them to be reasonable and relatively well supported by research: first, individual investors know more about the domestic market than they do about foreign markets, while mutual funds are equally well informed about all markets; and second, mutual funds differ in ability to generate abnormal returns. Next, they hypothesized that investors who know more about the domestic market could make better evaluations of mutual funds investing in domestic markets than funds investing in foreign markets. These would make mutual funds investing in domestic markets more attractive to investors. The challenge was to construct a manageable model to illustrate this intuition.

Dziuda and Mondria note that to evaluate managers’ abilities to generate abnormal returns, individual investors should look at funds’ past performance. Investors can easily identify strong funds among those that invest in the domestic economy: they perform much better than the economy.

It is much more difficult for investors to identify strong funds among foreign investments. Investors typically do not closely monitor the foreign economy and therefore cannot judge whether the performance of a foreign fund stems from its innate capabilities or from outside market forces.

Having less information about foreign markets, the investor’s ability to judge a manager’s capabilities is limited to comparing the results of this particular manager against the results of other managers. But investors independently learn the capabilities of domestic managers more quickly and easily than they learn the comparative abilities of managers in foreign markets. As a result, investors are able to allocate their domestic investment more efficiently, which makes that type of investment more attractive.

With investors able to allocate capital efficiently in domestic markets, home bias grows—along with the number of managers attracted to the domestic market and along with the diversity within the domestic market. Home bias not only persists, it is reinforced.

Symmetric Information
The confirmation of home bias has broader implications. Dziuda and Mondria show, for example, that mutual funds need not be asymmetrically informed for home bias to persist. They suggest that, unless proven otherwise, future research on mutual funds should begin with that assumption. More symmetric information at the level of the individual investor, however, could address some of the imbalance between foreign and domestic mutual fund investments. “Our results say that if individual investors had similar access to information about all markets, home bias would be diminished,” Dziuda says.

Yet Dziuda and Mondria also conclude that mutual funds in the domestic market should, on average, be of higher quality and generate higher excess returns. They hope to learn whether this prediction can be supported by future data. But all things considered, it seems investors are likely to maintain a significant preference for the comforts of home even while surrounded by a world of information.

Modeling Real Economies
Dziuda addresses a critical next step for this work: figuring out how the model reflects real economies. “Our model shows that there will be more investment in the domestic market than in the foreign market. The question is, however, which fraction of the phenomenon are we explaining? That is, if our model is an accurate description of the economy, what percentage of the investment will be devoted to the domestic market? Will it be higher or lower than in reality?” she asks.

“To check that point,” she continues, “one should assume reasonable values for the parameters of the model—such as discount factor or stock and asset volatility—and see what home bias the model predicts for these parameters. There have been many studies that tried to estimate what the reasonable values for these parameters are, and over time a certain consensus was established. Using those values, we obtain that our model predicts home bias close to the actual one.”

After running 10,000 simulations, their model showed that 73 percent of mutual fund investment was placed in domestic markets. Dziuda and Mondria concede that home bias predicted by their model is lower than shown in other studies, which showed aggregate home fund bias of 85 percent in the United States. But while their model might not explain what Dziuda and Mondria the call the “whole puzzle,” it does offer an explanation of why a large fraction of domestic capital stays so close to home.