����AV

In business, a specialist strategy can sometimes be riskier than a generalist one. Competing in only one industry leaves firms highly vulnerable to heightened income volatility, with extreme gains and losses, often alternating in quick succession. Innovative firms, whose business models are based on heavy R&D investments with uncertain returns, are especially affected by these fluctuations.

The tax system can exacerbate these specialist-specific risks through a principle called convexity—fat years are taxed at the full rate (e.g., 21 percent currently) without an equal rate of subsidy or assistance in years without a profit. As , assistant professor of accounting at Mason, explains, “Over two years, a diversified firm with a stable profit of $5 million per year would pay half the tax of a specialized firm that makes $20 million one year but loses $10 million the next.”

Policies that allow taxpayers to use current losses to get a refund of prior taxes paid (called a loss carryback) can help reduce this disparity. However, the Tax Cuts and Jobs Act (the 2017 tax reform) removed the ability for taxpayers to use carrybacks. The CARES Act subsequently reintroduced carrybacks, but only temporarily during the pandemic. While firms are still allowed to apply current losses to offset future income (called a loss carryforward), prior research shows that firms often do not fully utilize these loss offsets for a variety of reasons. As a result, convexity likely continues to play an important role in determining a firm’s tax liability.

An obvious opportunity for tax-wary specialists is to diversify their business in an attempt to stabilize overall financial performance. Classically, this is thought to be a key benefit of a conglomerate business model in the United States.

Wentland’s recent (forthcoming in The Accounting Review) puts this wisdom to the test, comparing the income volatility and tax burden of industry specialists and generalists over the period 1993-2017. Her sample, pulled from the Compustat corporate database, includes roughly 35,000 firm-year observations in the paper’s primary analysis. Wentland controls for a host of firm characteristics, including firm size, accounting for the fact that certain firms have more resources for limiting their tax exposure.

Wentland found that operating in an additional industry was associated with a 35 percent reduction in taxable income volatility and an 11.6 or 14 percent lowering of the tax burden for firms in her sample period—depending on whether taxes were scaled by assets or sales, respectively. “On average, I find that firms that operate in more than one industry have a tax savings of about $2.5 million per year,” Wentland says.

But the aggregate figures tell only part of the story. Not all diversified structures were equally effective as a tax hedge. When the different branches of the business had high cross-industry cash flow correlations—i.e., a GE financing branch supporting its automotive sales—there was little or no tax advantage conferred. Additionally, generalists that were very active in multiple countries were less protected from convexity, because tax jurisdictions may prevent losses in one jurisdiction from offsetting income in another jurisdiction. Similarly, firms that diversified through M&A (mergers and acquisitions) are prevented by the tax code from using an acquired company’s past losses to balance out the purchaser’s profits.

To isolate the role of the tax system characteristic (convexity) from other firm characteristics that influence tax planning, Wentland examined how firms responded to two temporary tax law changes expected to reduce convexity in the tax system: the Job Creation and Worker Assistance Act of 2002 and the Worker, Homeownership, and Business Assistance Act of 2009. As a short-term response to economic downturns, these regulations granted businesses greater access to tax refunds by extending the periods to which firms could use current losses to offset taxes paid on past income years. In doing so, they reduced the advantage with tax convexity for generalists amongst firms that were eligible, i.e., those with net operating losses while the policies were in force.

“It’s not obvious that if you follow the strategy, you always get a benefit though. You really do have to keep in mind the context of your business and how you grow this strategy,” Wentland says.

Wentland’s nuanced findings can help policymakers understand how the tax code may affect corporates’ willingness to engage in riskier strategies. Policies that act to reduce tax system convexity, such as loss carrybacks, level the tax playing field for specialists and more innovative firms, thereby encouraging them to continue down their potentially more volatile path. In the last few decades, policies regarding carrybacks have wavered between generous and prohibitive in reaction to the economic climate. Alternatively, policymakers could be more proactive. Wentland suggests they could ask themselves questions such as, “Do we want firms to be diversified and insulated? If all of our companies are specialized and we have events like the pandemic or economic recessions, what does that mean for our economy? How do we want to incentivize that risk-taking from tax policy?”

Rather than pushing for one policy direction or another, Wentland hopes her research will advance and refine the debate. She points out that risk-taking is sometimes not only desirable but necessary. Think of Amazon’s initial losses from investments in logistics and supply chains and Big Pharma’s investments with its COVID-19 vaccine and the role these innovative breakthroughs have played in the past two years. On the other hand, nudging businesses toward diversification could soften the overall blow of an economic downturn.

“Say, for instance, we’re headed into economic recovery and we really want people to be investing and taking risks,” Wentland says. “We might say, ‘Maybe we should subsidize risk. We should have a bit more carrybacks. We should do these things to make it okay to fail and keep going.’ On the other side of things, though, we might ask ourselves, ‘Are we potentially heading into a recession? Do we want to think about how we should be preparing to insulate ourselves?’ In that case, you may not want to offer that subsidy for loss.”