An academic paper awaiting publication poses a multi-trillion-dollar question: What are the most important determinants of U.S. investment in the equity of foreign firms?

Two of three favorite explanations for home bias do not withstand scrutiny, claim the report’s co-authors.

John Ammer at the Federal Reserve Board, Sara Holland at the University of Georgia, and David Smith and Frank Warnock at the University of Virginia base their findings on U.S. Treasury data used to calculate foreign holdings by U.S. residents. They relied on a confidential benchmark survey of company-level data conducted in 1997 by the U.S. Treasury Department and the Federal Reserve. Such granular detail on investor behavior is extremely rare and it cried out for the scholars’ attention. Conclusions shed overdue light and stand the test of time, they insist.

Despite such enormous stakes in foreign stocks, the co-authors cite a gap in research on what governs the level of U.S. investment in foreign firms. Their new paper makes a dent subject to newer data that they would welcome. The paper, entitled U.S. International Equity Investment, disputes two favored and long-held assumptions about disproportionate investment in foreign companies that cross-list their depository shares in the U.S. The first is that lower transaction costs did not explain the imbalance.

If the reduction in transaction costs motivated investors who wanted to trade stocks on the NYSE, most cross-listings would have been ADRs. Instead, underlying foreign securities received the bulk of attention from U.S. investors. This evidence rules out transaction costs as a primary impediment to foreign investments.

Institutions wait for firms to list in the U.S. before they invest, says David Smith, who directs the Center for Financial Innovation at the University of Virginia’s McIntire School of Commerce. When firms satisfy reporting requirements for cross-listing, investors instruct traders to buy shares where they are cheapest. More often than not, traders bypass ADRs for direct investments.

Common wisdom fails inspection again in the second of the favored explanations when it assigns a cross-listing effect to differences in legal systems. That was not a main draw, it turned out. “We didn’t find any evidence that ‘bonding’ to U.S. laws led to a larger ‘cross-listing effect,’” says Frank Warnock, the Paul M. Hammaker Associate Professor of Business Administration at The University of Virginia’s Darden School of Business.