The Agenda

Questions, Evasions, Etc.

Update: There’s a response at Daily Intel. I did indeed misinterpret some aspects of the post! Such are the perils of writing with spotty wi-fi on the train on little sleep. I definitely made some howlers — I didn’t see one of the charts the author posted, which is pretty sad. I owe my readers an apology, but you can’t say I didn’t warn you. 

But then the author writes:

Reihan also describes raising taxes on the rich as “raising marginal tax rates.” In fact, President Obama’s proposal would not touch their marginal tax rates, but would instead reduce the value of their deductions. This error seems symptomatic of his desire to engage with imaginary arguments rather than the ones actually being put before the country.

I didn’t actually reference the president’s proposal. Rather, I wrote the following:

My own view is that we might at some point need to increase the tax burden. One way to do this would be to phase out the mortgage interest deduction and replace the tax exclusion for employer-provided health insurance with a refundable credit. This isn’t as eye-catching as raising marginal tax rates on high earners, though it would tend to make the tax code more progressive. Such measures have the added advantage of improving allocative efficiency across the economy. What they won’t do is scratch the itch of those who believe that raising MTRs should take precedence over improving the cost-effectiveness of public service delivery.

There are lots of ways to raises taxes on the rich. The author suggests that I’m engaging imaginary arguments. I’d submit that there really are people who want to raise marginal tax rates! Moreover, reducing the value of deductions for people above a certain level of income is a way of increasing their effective marginal tax rates. The same is not true of eliminating a deduction. 

The following is from Alan Viard at AEI:

The effective marginal tax rate is the additional tax triggered by earning an additional dollar of income. This effective rate depends on the breadth of the tax base as well as the official marginal tax rate. For example, suppose that people always spend half of their income on apples and half on oranges. Also, suppose that the income tax has a 40 percent rate, but that spending on oranges is deductible. Although this tax has a 40 percent official marginal rate, it has an effective marginal rate of about 20 percent. In other words, its work disincentives are roughly the same as those of a 20 percent tax on both fruits, reflecting the combined impact of a 40 percent tax on working to buy apples and a zero tax on working to buy oranges. (As discussed below, despite the similarity in work disincentives, the 40 percent apple tax is markedly inferior to the 20 percent broad based tax in other respects, as it distorts consumers’ choices between the two fruits.) [Emphasis added]

There are, of course, lots of ways of looking at this particular issue. 

Do I think that my musings are of tenuous relevance? Yes, I believe that they are of tenuous relevance to the author of the original Daily Intel post. I do think that my musings — which is to say the questions I’m raising — might be of relevance to people who are interested in how our economy is changing. For example, the changing composition of the U.S. population seems at least slightly important to understanding what’s going on. I also think the sources of wage dispersion might be of interest. But of course some will disagree!

The author also talks about efficiency:

I argued that “Deficit reduction is (or is, at least, very close to) a zero-sum resource-allocation question.” Reihan replies by citing the possibility that the government might find ways to do more with less. That’s perfectly swell. Every president has tried it, none have enjoyed more than slight success, but maybe we can do better. But there’s no way to close the long-term deficit just by making government do more with less. Therefore, we’re back to the zero-sum resource allocation question — lower taxes for the rich means more cuts in other things. You can argue that lower taxes for the rich and lower Medicare benefits beats higher taxes for the rich and higher benefits, but you can’t wave the magic wand of more efficient government to make the question disappear.

As regular readers know, I’m an extremist on this question. The fact that Crossrail is so much cheaper than the Second Avenue Subway is, in my view, an interesting and important fact. The efficiency question is the ur-question of this blog, and the fact that many other folks hand-wave it away presents us with an opportunity. That said, I explicitly state later in the post that I’d be open to new revenues via the elimination of various deductions, which would of course mean higher average taxes on the rich. 


At New York’s Daily Intel blog, I come across a post that reflects the views of a large number of educated readers. I encourage you to read the post for yourself, as it is entirely possible that I’ve misinterpreted some aspects of the post. Here goes:

This is an economy that’s incredibly great for the very rich and pretty lousy for everybody else. If you consider this even mildly problematic, you have several responses. You can try to help workers unionize, or restructure the financial industry, or challenge the loopy way we pay CEOs more and more regardless of performance. You can look for even more radical solutions, which some of the protestors favor, though I would not.

Even if you reject all those options, at the very least, this is an important context in which to understand the debate over taxes. 

Let’s add more context. I’d argue that restructuring the financial industry is of the utmost importance. Of course, restructuring the financial industry might have no impact on wage and wealth dispersion. Indeed, it might actually increase the level of wage and wealth dispersion, e.g., if a more efficient financial system facilitate the rise of new entrants in a number of industrial sectors, creating new entrepreneurial fortunes and driving inefficient incumbent firms to bankruptcy. 

The author presents a series of graphs tracking average household income, and the change in the share of income earned by the top 1% and households by income since 1979. A good first step is to think through how the demographic composition of the U.S. population has changed over the intervening years. Has there been a sustained influx of less-skilled workers, with levels of educational attainment that more closely track those of southern European migrants of the early twentieth century rather than native-born U.S. residents? Have we seen a high level of family disruption, i.e., have we seen an increase in the number of U.S. households with children headed by single parents? Have we seen an increase in assortative mating, i.e., people with similar levels of educational attainment marrying each other? Are single-person households more common than they had been in 1979? Do we have reason to believe that there will be higher within-group wage and wealth dispersion in an older and more educated population, per the work of Thomas Lemieux?

Even if we correct for the changing demographic composition of U.S. households and changing family patterns, there remains a powerful underlying pattern of wage and wealth dispersion. We’ve discussed some of the sources of this underlying pattern, some of which can be traced to the incentive structures embraced by “digital organizations.” In the world of knowledge-intensive services, leading firms have tended to coalesce around a shared set of practices: workers are given a great deal of autonomy, and compensation tends to be performance-based. It is relatively common that two software programmers with identical paper qualifications will differ markedly in performance, and compensation at firms like Google and will tend to reflect this difference. This is particularly true when the worker in question has a set of “general-purpose” as opposed to asset-specific skills that can be deployed at a number of firms. Google, for example, sharply raised compensation for a number of programmers it feared might otherwise defect to Facebook or any number of high-growth start-ups, in part because equity can no longer be as powerful a draw for employees. 

So even if we limit ourselves to the “real economy,” there is a straightforward narrative in which wage dispersion within firms will increase. The work of Chad Syverson suggests that there is a parallel dispersion in outcomes among firms. Some firms, for a variety of reasons relating to organizational capital or capital productivity or managerial prowess, all of which are closely related concepts, can produce more stuff with the same raw inputs than other firms. Not surprisingly, the firms that do so are more likely to survive and thrive, and in the process drive competitors out of business, providing we’re dealing with a truly competitive marketplace, which is a big if. Public sector firms vary in productivity, yet firms in the public sector are often rewarded rather than punished for poor performance. 

The debate over executive compensation is an interesting one. The author might be channeling the story told by economic sociologist Rakesh Khurana, about the capture of corporate boards. Another story is that as the market capitalization of firms has increased, the tenure of CEOs has grown shorter and minor differences in CEO performance are more consequential. When we combine these two developments, it is fairly intuitive both that the services of CEOs who are considered highly effective would be deemed more valuable and that workers in the CEO labor market would seek more protection against downside risk. The CEO of a high-profile publicly-traded firm might fear that failure would have a lasting economic impact, and so she might negotiate a “golden parachute” before choosing to take on such a job. We also see that managerial compensation tends to be higher in closely-held firms, where the capture of boards isn’t quite as salient. The interests of owners and managers in closely-held firms tend to be better-aligned than in publicly-held firms, yet this hasn’t generally meant much lower compensation. All of this is contested, to be sure, but it is certainly worth thinking about. 

Raising taxes on the richest 1 percent would obviously not solve the skyrocketing inequality gap, but it would reduce it slightly, while imposing far less pain than the alternatives.

Raising taxes would presumably have no income on market income inequality. Rather, it would have some small impact on post-tax inequality. This is one reason why we tend to look to post-tax-and-transfer inequality to get a good understanding of the shape of the economy. Yet we’re presented with charts showing average household income before taxes. Post-tax-and-transfer charts wouldn’t give us a markedly different picture, but they would give us a somewhat different picture. One of the more egregious scandals of our time has been the ballooning cost of various tax expenditures that disproportionately benefit the affluent. We have also seen expanded eligibility for transfer programs including the EITC, food stamps, and Medicaid, among others. We could both think that this is a very good thing and that it would be useful to factor the value of these benefits into a survey of post-tax-and-transfer inequality.

Deficit reduction is (or is, at least, very close to) a zero-sum resource-allocation question. The interests of the richest 1 percent differ from those of everybody else.

This is actually very important. If we believe that the public sector is a perfectly fixed entity that can’t get any more efficient than it is and that must necessarily expand in size as, say, the population ages, this might make sense. There is, however, good reason to believe that public sector firms vary in quality quite dramatically, from jurisdiction to jurisdiction and even within jurisdictions. The fact that less-affluent people are more likely to draw on public services means that raising the quality of public services is particularly important to the less-affluent slice of the population. This leads us to the question of which strategies are best for improving the performance of public sector firms. One theory is that fiscal discipline can create a powerful impetus for organizational discipline, and for improved service delivery. Unfortunately, this only applies when the firms in question are legally permitted to become organizationally disciplined, i.e., to reform work rules, compensation structures, etc., to best deploy a given set of resources to tackle a given set of problems. Unfortunately, public sector firms are more likely than private sector firms to have their hands tied in this domain, which all but guarantees that productivity gains will be severely limited.

All of this is to say that deficit reduction could actually be a spur to broad-based productivity improvements, provided public sector managers are given the autonomy they need to restructure their organizations. Suffice it to say, this is a controversial issue. When the potential for organizational reform is severely constrained, deficit reduction is hard to reconcile with maintaining a given quality of service. 

It’s interesting to note how context clarifies and changes things. 

If you search deep enough into the conservative liturgy, you will find genuine arguments on behalf of holding harmless the top 1 percent. There’s the argument that higher taxes on the rich will impair growth (an argument belied by the Clinton years, but never mind.) There’s the argument that the rich, as a matter of fairness, should not be forced to pay higher tax rates than the middle class. Instead of making these arguments, though, the right is evading the question altogether.

I’m certainly not an expert on the conservative liturgy. What I will say is this: I don’t think it’s evading the question to first ask where the money is actually going and how it is actually being spent before we discuss raising taxes on anyone. Perhaps we should take the existing tax burden and shift it in various ways, e.g., by creating a progressive consumption tax along the lines supported by Robert Frank on the left and Alan Viard on the right. As Matthew Slaughter has recommended, we might restructure the payroll tax so that it is less burdensome on working and middle income households. There are any number of things we might do. But the real issue, for me at least, is whether or not government is functioning in a cost-effective way.

My own view is that we might at some point need to increase the tax burden. One way to do this would be to phase out the mortgage interest deduction and replace the tax exclusion for employer-provided health insurance with a refundable credit. This isn’t as eye-catching as raising marginal tax rates on high earners, though it would tend to make the tax code more progressive. Such measures have the added advantage of improving allocative efficiency across the economy. What they won’t do is scratch the itch of those who believe that raising MTRs should take precedence over improving the cost-effectiveness of public service delivery. 


The Latest