Limited effectiveness and high distortionary effects are the reasons why most economists historically have not favored corporate income taxes.
"The first argument against corporate taxes is that they aren't that helpful. The second is that they are very, very distortionary, and you want to rely on them very little," said presenter Alan Auerbach. There was some support for that view in the afternoon session of “The Facts About Taxes,” but Auerbach maintained that the actual picture stands as much more nuanced than corporate taxes being simply good or bad.
Auerbach stressed that talking about capital income taxes means talking about a whole host of different sorts of taxes. "Capital income" can refer to many things— rents and quasi-rents to old capital, labor income disguised as corporate income, compensation for taking risks among them. All of these have their own independent effects when taxed.
Moreover, since policymakers tend to tinker with these elements over relatively short time scales, businesses face "dynamic inconsistency" in what to expect from tax policy, making it difficult to plan ahead for the effects taxes will have operations decades down the line.
Auerbach noted that conventional thought has viewed a system taxing labor income but not capital and a system only taxing consumption as essentially equivalent. Again, Auerbach said, that isn't the whole picture. In the transition between systems, wealth accumulated and taxed under a labor tax would then be taxed again when spent under the new consumption tax. Auerbach said that these transition costs would be more significant than consumption tax proponents believe.
“You see huge differences in long-run GDP and economic efficiency to the point that the differences in shifting between these two tax systems is far bigger than the difference associated with an efficiency gain associated with not taxing capital." Thus, it may make more sense to ease capital taxation within our existing system, rather than radically overhaul the tax system with a shift to a consumption tax.
Despite historical concerns over inefficiency, Auerbach proposed several lines of reasoning on why we might continue to tax capital income today. Taxes could be used to control the conversion of capital income into labor income; to regulate inefficient pooling of risk in private markets; or to limit the mobility of fixed rents. By tweaking or removing deductions of interest payments, capital income taxes can incentivize companies to save and invest using equity instead of debt.
But, like income taxes, capital income taxes can be structured to maximize the labor supply for a country's economy as well. Auerbach asserted that capital income taxes can assure that, over time, people of high ability don't accrue such high savings that they lose the incentive to work during economic downturn, at the moment that the need for labor that can create growth is greatest.
However, Auerbach was careful to note that we can’t evaluate the impact of taxes on people’s lives exclusively based only on long-term effects. "You're talking about asking people to put income that could be enjoyed now for 40 years. That's a tough decision for people,” said Auerbach.
Structuring capital income taxes to keep skilled workers in the labor force also results in some fairly counterintuitive tax policies, like raising taxes at the precise moment that people are out of work in a rough economy, to force them back into the labor market. "You can achieve this outcome even with capital income taxes that average at zero if the tax rates vary based on the state of nature—on the the economic conditions."
Moreover, such tax policies rely heavily on the presumption that government will always act in the best interest of the people it governs. "When you're thinking about behavioral economics or behavioral public finance in the context of tax policy, you are at some level talking about paternalism," said Auerbach. "And that should make some people feel uncomfortable."
Auerbach concluded by stressing that in addition to considering what we're taxing and why, architects of tax policy need to deeply consider how the taxes they impose will be collected and enforced in an increasingly complex international economy.
"It's really hard to know what the optimal tax system would be, given the number of margins to consider in an international context," said Auerbach. "It's not just the [relationship between] government and the taxpayer, but the [relationship between] government and other governments as well."
With taxes levied on corporations across borders, as well as across generations of people, knowing how those taxes will be enforced will be critical to being able to predict their long-term effects.
Calling Auerbach's paper as "an excellent overview of the existing theories of capital taxation," Princeton University's Mikhail Golosov sought only add numbers to some of the theory laid out in the presentation. "I want to lay out some quantitative predictions—how big the effects Alan talks about are."
Golosov focused on one main question related to how taxation of saving affects individuals over their lifetimes: what parameters should we use to determine the size of taxes, and how much do we know about those parameters from empirical evidence?
Within the model he employed in his analysis, individuals of varying incomes and with varying preferences live out their lives, occasionally suffering individual shocks to their productivity. In this context, the goal of tax policy is to redistribute income between all individuals in the economy, and to mitigate the effects of those shocks in their lives. Given that framing, taxation is just a means of providing insurance for those worst affected by shocks to their ability to work and earn income on their own.
Whether because they inherited more wealth or because they're particularly productive entrepreneurs, Golosov asserted high-earning individuals earn a disproportionate amount of the capital income in this analysis, meaning they pay a disproportionate share of the taxes on capital income. Golosov said that the data backs up the theory guiding how these types of taxes affect individuals in the actual economy, but further research is required to empirically verify more of such parameters.
Discussant Wojciech Kopczuk of Columbia University dovetailed Golosov's analysis. "The reason we want capital income taxes is because we want insurance," said Kopczuk. "But if we tax them too much, we create disincentives." While social value exists in offering that insurance, he pointed out that the research over the past forty years suggests that offering insurance via capital taxation is still likely not optimal compared to other policy alternatives.
Kopczuk stressed the point that capital taxation much like the broader economy–has changed considerably since the 1970s, especially when we consider the complexity of financial instruments available to corporations today. "The ability to reclassify debt as equity, or the other way around, likely has increased over time," said Kopczuk. "This presents constraints that were not at the forefront of tax policy forty years ago."
The complexities presented by rapid financial innovation, the rise of multinational corporations, and entrepreneurs whose contributions aren't clearly delineated as labor or capital income all pointed Kopczuk toward one conclusion: a consumption based tax would be easier to work with in many ways.
While the transition costs certainly exist, taxing wealth as it is consumed, regardless of on what, would require much less accounting of distortionary effects present throughout the multitude of categories of capital income, said Kopczuk.
Casey Mulligan, a professor of economics at the University of Chicago, agreed with other discussants that finding the appropriate components of a heterogeneous mix of capital assets to tax has become increasingly and exceedingly complex. His solution? Don't tax capital assets at all.
"The rates of income earned on these types of assets are entirely different, by a factor of three," said Mulligan. "The number one reason for that? Taxes." Failure to tax all capital uniformly creates significant deadweight losses, according to Mulligan's own work.
Mulligan argued that we saw the consequences of this policy failure play out during the financial crisis. "Would the housing boom have been the same if we hadn't been overbuilt due to taxes?" Rather than treating different types of personal and corporate capital differently, he maintained that a much easier policy alternative exists. "Zero is a pretty easy tax figure to keep track of."
Roger Gordon of the University of California, San Diego rounded out the discussion by purposely taking a more controversial stance to stimulate debate about making substantial changes to the corporate tax system.
First, Gordon noted that looking at the effect of taxes on savings over a life-cycle model is a bit flawed in more contemporary literature. People save not only for retirement in the long term, but for job losses, college educations, and other shocks on shorter time scales where the effect of tax policy is smaller. "Behavioral models say that we're oblivious in this case to changes in tax law," said Gordon.
He agreed with Mulligan that "the effective tax rate varies hugely by the type of asset." Having fully taxable income from some assets with deductible interest and exemptions for others makes for policy that is complex, unwieldy and distortionary.
Gordon argued that a complex corporate tax code results in all sorts of tax arbitrage to avoid paying high personal income rates, particularly in higher income brackets. For instance, people might borrow to buy more lightly taxed assets, deducting the interest payments in the process. Bond base on that debt are bought up by tax-exempt funds and low-income bracket investors, which introduces unnecessary instability and inefficiency costs into the economy.
Moreover, as other discussants mentioned, income shifting makes differentiating between personal income and corporate income a complex exercise. Gordon suggested that correcting that behavior would require bringing the corporate tax rate closely in line with the highest personal income tax rate, a nearly 10 percent increase that would certainly have its own efficiency costs.
These complications, coupled with the complexities of extracting taxes from multinational corporations that shift their capital all over the world to minimize their tax burdens, lead Gordon to also favor a consumption tax over the existing corporate income tax structure. He asserted that it would eliminate many distortions, the need for corporate taxation altogether, and the need for tax repatriation. Instead, everything would be captured the consumption tax.
In discussion, this last point was heavily debated by others in the audience, including Auerbach himself, who reiterated the point that making such a policy change is not as simple as throwing a switch, both politically and logistically.
Former Obama economic advisor and Chicago Booth professor Austan Goolsbee raised the question of whether the bickering over deductions would simply shift to bickering over what counts as consumption. That said, Auerbach agreed that the solution to capital fleeing abroad may very well be a consumption tax-based system, since it matters only if people, not paperwork, move abroad.