When a U.S.-based company expands overseas, it increases the number of jurisdictions in which it earns revenue, and its earnings are taxable in both the foreign jurisdiction and the United States. However, the U.S. tax is not due until the foreign subsidiary pays a dividend or repatriates its earnings to its U.S. parent. My research examines the phenomenon of “trapped cash,” in which companies decline to repatriate assets held in foreign countries because of the relatively steep tax rate they would pay to bring those assets back into the United States.
Recent years have seen an unprecedented buildup in foreign cash balances by U.S. multinational firms—in one estimate for about 600 of the largest U.S. multinationals in 2012, foreign cash equaled $588 billion, or 60 percent of their total cash balances. Members of Congress have expressed concern over this buildup because it potentially represents foregone tax revenue—the allegation being that firms are escaping taxes by keeping their money overseas—and because these balances represent money that could be spent to buy raw materials, manufacture products, or pay workers or shareholders in the United States.
Of course, not all cash held overseas is trapped. Companies also hold cash overseas, for example, because they intend to use it to fund operations in foreign markets. Not all cash held overseas is taxable either: loans or capital contributions from the parent company represent foreign cash that is not taxable in the U.S. Thus, the political discourse about whether to change U.S. tax laws so that more cash might instead be repatriated is largely speculative, since U.S. companies are encouraged to declare how much cash they hold overseas, but are not required to divulge their motives for keeping the cash there. In a new working paper with Wayne Nesbitt of Michigan State University, we develop a way to estimate the likelihood that a given company actually has trapped cash.
To estimate the likelihood of trapped cash, we take advantage of a one-time event: the American Jobs Creation Act of 2004 (AJCA), which offered firms a one-time tax rate reduction on repatriated foreign earnings, reducing the effective tax rate from 35 percent to 5.25 percent.
For our study, we start with a list of all firms that appeared in the S&P 500 between 1999 and 2010, and examine these firms’ financial statements from 2004, 2005, and 2006, the years during which firms were allowed to repatriate at the lower rate under the AJCA. Basically, we assume that if a company repatriated cash without borrowing money to fund the repatriation, the company held trapped cash prior to repatriating it. After separating our sample into firms that had trapped cash and those that did not, we looked at several factors that we predicted would differ between the two groups.
One factor that differs is borrowing: we find that the less credit-worthy a firm is, the more likely it is to have trapped cash. Banks are more willing to loan money to firms with better credit. Some companies actually borrowed money in 2004 to repatriate it at the lower tax rate; by definition, these companies do not have trapped cash. If they did, they would not have needed to borrow money because they would have had cash to repatriate. We provide examples of companies that use debt as part of their tax-planning strategies, and others that would like to but are unable to. For example, in 2013, Apple borrowed $17 billion to help fund a stock buyback rather than tapping into their offshore cash pile of approximately $100 billion, thereby saving the firm about $9 billion in taxes. In contrast, Avon decided to bring back cash held abroad after the company unsuccessfully tried to renegotiate the terms of some debt, admitting this was not an ideal outcome at least in part because of the taxes.
In addition, we find that firms with robust research and development (R&D) programs are more likely to have trapped cash, possibly because adverse economic shocks are more costly for these firms, because they have more intangible assets—making it easier to shift profits to low-tax jurisdictions—or because this is capturing growth. More capital-intensive firms are less likely to have trapped cash because they tend to reinvest foreign earnings in operating assets in the markets where those earnings are generated. Similarly, firms with high foreign growth rates are less likely to have trapped cash, primarily because earnings generated in foreign markets will be re-invested to fund growth.
One might assume that firms that hold cash overseas to avoid paying taxes would also be likely to hold large cash reserves in “tax havens”—jurisdictions with especially low tax rates. Actually, we found the opposite, perhaps because firms that operate in tax havens are generally more savvy and proactive in their tax planning, structuring their operations to enable tax-free access to cash instead of having it trapped overseas.
Thus, we find that four factors—credit-worthiness, R&D spending, capital intensity, and foreign growth rate—are associated with trapped cash. We validate our model using two separate methods: examining the percentage of debt firms used to fund repatriation and examining the amount of cash held overseas in 2010 through 2012—a new requirement for firms to disclose. Firms that we find are likely to hold trapped cash, based on our analysis of the data from 2004-06, hold substantially more cash abroad than comparable firms that we find are unlikely to have trapped cash based on the 2004-06 data. These validation checks provide comfort that our methodology represents a parsimonious way to identify firms with trapped cash, and it can be applied to large samples of firms regardless of whether they disclose their foreign cash holdings.
Lastly, we examine the relationship between trapped cash and firm value, and find that on average that relationship is negative. However, the negative relationship is driven by poorly governed firms. Thus, investors apparently view trapped cash as a resource that could be put to more efficient use, but realize that this might not happen when corporate governance is weak.
In summary, our study develops a new method for estimating the likelihood that a firm has trapped cash—an important contribution because it is difficult to have a reasoned political debate about this issue without having a good sense of the size of the problem legislators are discussing. Although we take advantage of a one-time event to develop our measure, validation checks indicate that our method is widely applicable. Also, our finding regarding firm value suggests that trapped cash is indeed a problem, and not just evidence of smart tax planning on the part of companies with large overseas cash reserves. Our study substantially furthers our understanding of the phenomenon of trapped cash—and Wisconsin School of Business students are already benefiting from this cutting-edge research. I presented these results to the students in the MAcc program last month to help them understand how multinational firms use tax planning strategies including holding cash overseas—and how research ties in to the content of their education and the scope of their work as practicing accountants.