Income versus consumption taxes

In a commentary last fall in the NY Times, Robert Frank makes the case for consumption taxes to replace the income tax in the US. Yet, while this sounds revolutionary on first reading, what Frank is describing is essentially the Canadian tax system:

Under such a tax, people would report not only their income but also their annual savings, as many already do under 401(k) plans and other retirement accounts. A family’s annual consumption is simply the difference between its income and its annual savings. That amount, minus a standard deduction — say, $30,000 for a family of four — would be the family’s taxable consumption. Rates would start low, like 10 percent. A family that earned $50,000 and saved $5,000 would thus have taxable consumption of $15,000. It would pay only $1,500 in tax. Under the current system of federal income taxes, this family would pay about $3,000 a year.

As taxable consumption rises, the tax rate on additional consumption would also rise. With a progressive income tax, marginal tax rates cannot rise beyond a certain threshold without threatening incentives to save and invest. Under a progressive consumption tax, however, higher marginal tax rates actually strengthen those incentives.

I had thought the US system was much like the Canadian one, in allowing deductions for retirement savings (401k there, RRSP here). And of course there is no federal consumption tax in the US. In Canada, generous deductions for RRSPs are such that only families with incomes in six digits have a chance at maxing out that deduction.

As tax economist Jon Kesselman points out this means we already have a largely consumption tax based system, some of it administered through the income tax system, some at the cash register. Here is Kesselman in recent testimony to the House of Commons Finance committee (no link as he sent me these remarks electronically):

[A]n economically efficient and socially equitable taxation system hinges on its overall structure and design – not on its overall level.  Other countries have applied taxes at levels significantly higher than Canada while still attaining very good levels of productivity growth and enviable standards of social programs. The key to achieving an efficient and growth-oriented tax system – whatever the desired overall level of taxes and size of government – is to shift the taxable base further toward consumption and labour income and away from capital and investment income.

… Canadian governments at both the federal and provincial levels made significant progress in pushing the tax system in the desired direction, and further constructive changes have been made in the four years since [2004].  Some examples include large reductions in corporate income tax rates, reduction and elimination of corporate capital taxes, expanded allowances for depreciation of business capital, modest reductions in personal tax rates, and the introduction of Tax-Free Savings Accounts.  All of these measures move the tax system further toward consumption-based levies and away from income- and capital-based levies.

That said, Kesselman’s argument is a theoretical one rooted in the flawed “savings leads to investment” model. But as an empirical matter, it is far from clear that such shifts have any demonstrable economic benefit (ie any empirical linkage to higher and sustained economic growth or productivity rates). There are huge distributional issues that are neglected in this perspective, and these arguably have a much more profound impact on economic outcomes. Any changes that further expand RRSP ceilings (or new accounts like the TFSAs) tend to be of benefit only to the wealthiest families. Meanwhile, most ordinary families have little or no savings.

Anyway, coming back to Frank’s piece, it is interesting to note that in addition to desires by Paul Krugman and others to emulate single-payer health care on the Canadian model, here is another economist wanting to emulate the Canadian tax system (though we get not credit there). The think is, I doubt it will have the impacts Frank is claiming:

 If a progressive consumption tax were phased in gradually, its main effect would be to shift spending from consumption to investment, causing productivity and incomes to rise faster.

The biggest benefit would be from higher overall taxation levels if used to finance single-payer health care; that is, the gains are on the spending side not through a more “efficient” tax system.


  • I generally favour the idea of taxing consumption, rather than the earning of income, but NOT so as to encourage faster economic growth or even increased labour productivity (although the latter is not necessarily a bad thing). The point, for me of having the incidence of taxation fall on consumption, is to encourage conservation of natural capital, and to enhance RESOURCE productivity. As any engineer knows, and as economists and management accountants also ought to know (but sometimes forget), we need to economize in our use of the scarcest factors of production. The scarcest factors of production, in a long-run dynamic sense, are those with the least-elastic long-run supply functions. In the long run, I suspect natural capital assets have the least elastic supply functions, because they have no close substitutes and we generally don’t know how to reproduce them. Even when we appear to have a steady supply of some natural capital input, or appear to have an ample stock on hand to perform a given “fund” function that only natural capital “funds” can provide, we are quite likely to, in fact, be misstaken. This is because our understanding of biology and ecology is still quite rudimentary, and because the “supply” of natural capital to the human economy is so often merely the mirror image of the depletion of natural capital stocks in the wild part of the global ecosystem, that we don’t think is important, and so we don’t generally monitor.

  • The Post Keynesian literature suggests that problem with the Savings –> Investment is much more serious than its empirical difficulties.

    It is also wrong institutionally — in the facts of how lending and savings work in the modern-day monetary economy of a sovereign nation — and theoretically.

    The upshot is that the causality is the other way around : investment calls forth money which then circulates in the form of wages, consumption, and so on. That’s just one of the many reasons why investment is the linchpin to understanding Keynesian economics and economic cycles — without those animal spirits compelling businesses and consumers to invest and in so doing, create new money into the economy, the whole machine grinds to a halt (short of, of course, a concerted effort at deficit spending by the sovereign or strong balance of payment surpluses).

    The repayment of debt extinguishes money, something known as the reflux principle (interestingly, the reflux principle has a history going back to Thomas Tooke, a 19th century political economist. In the modern day context, Marc Lavoie has written extensively about this, as have numerous other Post Keynesian economists.

  • I am a huge Robert Frank fan, but share the skepticism expressed by Marc and Arun. The NY Times op-ed proposes, “replacing federal income taxes with a steeply progressive consumption tax.”

    Other things being equal, taxing consumption instead of income is regressive. To me, the key words in this proposal are “steeply progressive.”

    A 100% marginal tax rate on income is almost unthinkable because it would essentially eliminate the incentive to earn income (above the threshold for that rate). But Frank envisions a 100% top marginal rate on consumption. This approach would either deter the super-rich from consuming beyond the threshold (to the benefit of everyone else) or force them to pay hefty tax bills on excessive consumption (again to the benefit of everyone else). The progressive argument for consumption taxation is that it permits far higher tax rates at the top end.

    However, top Canadian and American income tax rates could be increased without meaningfully reducing the incentive to earn income. For example, a 50% top marginal income tax rate is quite plausible. Someone making a million dollars over the top threshold would pay $500,000 of tax on this income and could allocate the remaining $500,000 to consumption. Under Frank’s scheme, this person would pay no income tax, but would pay $500,000 in consumption tax if they still spent $500,000 on consumption above the threshold.

    Either way, the million dollars in question finances $500,000 of consumption and $500,000 of taxes. The possible advantage of a 100% consumption tax over a 50% income tax is that the former might induce the millionaire to consume less and save instead. However, the social benefit of more private saving is far from clear. Despite the lack of household savings, corporate Canada is already sitting on a huge pool of savings because its internally-generated funds exceed its capital investment. (Tom Palley argues that saving is not the real issue in the US either.)

    Finally, under Frank’s proposal, increased saving would reduce tax payments. This loss of public revenue obviously entails a social cost, which might well outweigh the anticipated benefit of more saving. If the goal is “increasing federal revenue substantially,” why not introduce higher tax rates on consumption at the top end while retaining (or even raising) existing tax rates on income that is saved?

    It seems to me that Frank makes an excellent case for higher taxes on consumption at the top end, but not much of a case for completely exempting savings from taxation.

  • Erin wrote “… However, the social benefit of more private saving is far from clear ..” with which, in and of itself, I don’t disagree. However, the social benefit of more consumption by the wealthy (apart perhaps from the benefit of “trickle-down” effects to the serving classes) entirely escapes me. Not only does high consumption by the wealthy crowd out (at least to some extent) basic consumption of goods in inelastic supply (like housing and certain kinds of food) by the less wealthy, but it also preempts consumption of the non-renewable resource component of that consumption (by the wealthy) by future generations of human beings, as well as directly diminishing “eco-space” for other speciesm now and in the future. I have to say that I find it a bit disappointing that so many progressive economists seem to find so many reasons to support the income tax status quo , when fundamental reform is so badly needed.

  • To be clear, I agree that there would be a social benefit from curbing consumption by the super-rich. I also think that there is a social benefit from public revenue.

    I could support a high consumption tax at the top end because it would tend to limit excess consumption and generate more public revenue. I am inclined to oppose exempting savings from taxation because, although this approach might reduce high-end consumption, it would also reduce public revenues.

  • How easy would it be for the rich to hide private consumption as productive consumption. I can imagine a lot more liquid crystal displays and yachts being purchased by the company.

  • I’m not sure excessive consumption by the rich is a major problem. I’d almost rather the rich spend their money on yachts and OLED TVs so the money really might get into the hands of the middle and lower class.

    If they invest it instead it will only make them more powerful and richer and given them even greater control over the country and its people.

  • I’m not opposed to more public revenue out any particular market fundamentalism, but more public revenue is not, in and of itself, the solution to all problems.

    Our wonderful federal government wastes vast amounts of the revenue it already takes in – e.g. subsidizing the economically pathetic east coast seal hunt for example and then grandstanding for impressionable voters by getting the coast guard to arrest Paul Watson and the MV Farley Mowat, to quote two recent topical examples. It is also, to my horror, planning to pay hog armers to kill their now overstocked pig populations, having no doubt encouraged them to over-breed just a few years earlier, because some demented government economist figured that would increase GDP.

    The provincial government where I live, in BC, is hardly any better. It subsidizes oil and gas exploration, and a natural capital-destructive open-net-cage fish-farming industry, and wants to use public money to promote hunting, because not enough of us seem to want, of our own accord, to help kill off our diminishing wildlife populations.

    There are undoubtedly some public goods that are insufficiently supplied (like wilderness conservation or education in ecological economics to give two examples close to my heart) but simply raising more public revenue will not necessarily ensure that those public goods that are actually needed, are in fact supplied.

    In my opinion, market fundamentalists certainly need to recognize the pervasiveness of ecological externalities and the empirical frequency of market failure, not to speak of the inherent species-ism of the legal system that underpins the only market institutions we have ever tried out.

    Partisans of more public revenue at all costs, on the other hand, surely need to pay more attention to the limits of collective decision-making. Some respect for Kenneth Arrow’s Impossibility Theorem would be a good starting point.

    As was famously said, (Electoral) Democracy is the worst of all possible systems, except for all the others. While that doesn’t mean we should switch from democracy to absolute monarchy, dictatorship or oligarchy, we should surely not naively assume electoral democracy to be any comprehensively more perfect a system for allocating economic goods than the equally all-too imperfect market.

  • Notice that Arrow’s theorem applies to all aggregate preference ranking systems…including markets. There is some irony that every time neoclassicals attempt to resolve a contradiction within their own universe they do so by specifying criteria which resolve “fundamental” flaws in revival paradigms…capital controversy and the labour theory of value, for example; and when they create a paradox for non-market based forms of social choice they inadvertently sabotage market based forms of collective decision making.

    From a policy point of view Arrow moves us no farther along than Bob Marley: “you can fool some people sometime but you can’t fool all the people all the time and now you see the light…hey stand up for you rights….”

    I need to spend more of my time formalizing nuggets of folk wisdom.

  • “Notice that Arrow’s theorem applies to all aggregate preference ranking systems… including markets”.

    I guess that is exactly my point. There is no perfect way to divine “the collective good” and there is no unambiguously optimal distribution of wealth. Markets can do an efficient job of allocation, under the right conditions, but those “conditions” are all-too frequently not present, while markets do an ethically blind and often quite indefensible job of distribution. And, in addition, both markets AND electoral democracies do a very myopic job of keeping the macro-economic scale of the human economy within the biological carrying capacity of the planet.

    That doesn’t mean, surely, that we give up trying to improve what we find around us, but be should move humbly and cautiously, and we should be prepared to retreat when things don’t seem to be working out too well. Moving too aggressively towards a consumption-based system of tax without observing the distributional impact might well be misstaken. On the other hand, sticking blindly with our present heavy reliance on income taxes in the face of their high deadweight costs, and the strong, but perverse incentives they create for employers (even public sector ones) to economize on labour rather than to economize on entropic throughput, seems to me to be facing in the wrong direction. Even if we are doing so out of laudable concern for equitable distribution, it is still the wrong direction. Perhaps we need to deal with wealth distribution more directly – one could in fact argue that a significant one-time redistribution of wealth from rentiers to the working poor would actually be a macro-economic stabilizer in today’s credit markets. Or otherwise, should we not try to explicitly design distributionally-progressive consumption taxes ?

    Trying to score a few more years of free lunch at the expense of other species and those not yet born (of our own species and others) hardly seems an ethically respectable course of action, and it is really not sustainable beyond one or two more generations.

  • Michael,

    Can you provide any evidence of the “deadweight costs of the income tax system”. I already pointed out in the original post that the income tax system is in face a consumption tax system for the vast majority of tax filers.

    No unambiguously optimal distribution of wealth? I guess it is a question of values, but a grossly unequal distribution of wealth fails many important tests in terms of basic freedoms and opportunities, crime, health, and even economic performance.

    Government wastes vast amounts of money? There is bound to be some waste but most public money is spent on people delivering public services. As Jim points out in a recent post, there is much more waste in the private sector than in the public. It’s just that the media do not put their microscope on corporate waste.

  • I am skeptical with respect to the claim that income taxes are the cause of labour saving in capitalism. Given no maximum rate of return exists and given no maximum on income shares exists, the drive is to maximize profits which is determined in a relative sense. In a world where 20% was the average ROI companies would seek to beat that average by introducing labour saving techniques. And that holds at 4% as well. Prior to personal income taxes there were a mother-load of labour saving inventions and techniques of production. Income taxes basically did not exist when Ford pioneered the assembly line.

    The drive to introduce labour saving techniques is intrinsic to capitalism. Notice that productivity growth in China is higher than in Canada and wages are much lower. Cheap labour is good…less cheap labour is even better. (And, yes I know China has a lower level of productivity so it easier to increase productivity but that is not at issue here.)

    Do income taxes exaggerate this drive to introduce labour saving? I doubt it. If tomorrow firms were not required to “pay” income taxes they would still compete and try to beat the average ROI which would lead them to attempt to save on labour among other things. The world would look the same after the initial period of adjustment.

    And corporate waste: Think of the untold billions wasted chasing after every new management fad. The mind boggles.

  • Deadweight costs of income tax system ?

    Has anyone counted the costs of compliance (by taxpayers) and of enforcement (by government) ?
    Alan Durning estimated that in the US, 9c on every government tax dollar raised from income tax is spent by taxpayers complying with the law, and that doesn’t include enforcement costs borne by the IRS.

    Durning also claims that as much as 15% of potential income tax is evaded. That wouldn’t surprise me.

    I must admit to not have any Canadian data at my fingertips, but I believe that there have been estimates that tax compliance and collection may consume as much as 20% of potential GDP in any industrialized country.

    I don’t (in some ways regrettably) get paid to research and study tax policy as an economist, so I don’t really have the the time or resources to dig much deeper; but I do presently make my living as an accountant, largely out of helping taxpayers with compliance, and I must admit to not feel particularly “productive” (in a social sense) doing so. And I am just one accountant amongst thousands, many of whom serve more clients, and bill their clients far more than I do !

    And, interestingly, from a distributional point of view, I help a good number of modest income taxpayers who feel quite incapable (and afraid) of preparing their own tax returns. I don’t charge them that much, but my fees to them still represent a regressive “tax” of sorts. I can’t often save them much, since as others on this site have frequently pointed out, you need a considerable degree of creditworthiness, an appetite for risk and/or some cash surplus on hand, which three things many of these clients lack, to get any meaningful tax breaks. Yet the complex monster of the income tax system, which I had no hand in designing, intimidates them into seeking professional help, for which they have to pay, since I can’t afford to donate more of my time than I do already (mostly to environmental NGO’s).

    I am not saying that there is any costless way to raise public revenue, nor that we don’t desperately need many public goods the market will never provide. I am just saying we human beings have some really big ecological-economic problems looming ahead of us. Those problems have to do with stocks of natural capital assets and their rates of depletion. I am also saying that the tax system is probably the most efficient public policy toolkit to address those really big problems, because the concept of “taxes as the price of civilization” has a certain legitimacy, (even to many conservatives) – a legitimacy that micro-level regulatory consumption restraints do not. As a result, I think there is a lot more to the tax policy debate than just discussing its distributional impact amongst human citizens over the relatively short haul. And as a result also, I think there must be some room for moving the incidence of taxation away from the “adding of value” and on to the extraction and consumption of “that to which value is added”.

    The debate about the proper relative shares of the economy to be left to the market and to be managed in the public sector is a bit of a side issue really. I do see allocative efficiency in market outcomes that political decision-making, even when democratic, struggles to match. I realize that my view may be a minority position among bloggers and commentators on this website, in that regard. But I am quite explicitly not a market fundamentalist.

    I said quite clearly that I recognize that market systems are entirely amoral about wealth distribution, despite the fact that if wealth distribution gets too unequal, market systems cannot, in fact survive for long. I also hold positions that are in some way much more radical than many self-styled “social democrats”, in that I have serious doubts, on ethical grounds, about the legitimacy of private property in land (which amounts to a micro-ecosystem), animals and plants. I do however, believe strongly in private property in the abstract value added by an individuals’s own intelligence, labour, and entrepreneurship, to inanimate matter. I realize this is a difficult philosophical position to turn into practical economic ground rules, because even a single shovel-load of earth, or a bucket of water, contains living things. But I do think this line of thought is worthy of further consideration and discussion, in the broader economic policy debate.

    As to incentives to economize on labour inputs, surely it is a question of what you compare with what. Certainly, any entrepreneur seeks to minimize his labour costs, and would do so with or without income tax. But compare an economy which had no tax on wage income, with one that was identical in every respect except that it had a progressive tax on labour income. In which one would a new potential employer preferentially locate ? And in which economy would workers (if they were mobile) choose to locate ? And if there were employers of equal sized labour forces in each economy at a given time, and if wages rose 10% in both simultaneously, which employer, at the margin, would look hardest at alternatives that allowed him/her to reduce his/her labour force ?

  • “As to incentives to economize on labour inputs….”

    But then the culprit is the possibility of regulatory arbitrage not income taxes. Differential income tax rates between jurisdictions creates the potential for regulatory arbitrage. But we could just slap a tax on such cases and level the playing field. Or an international accord on some minimum level of personal income taxes.

    But all of this is very different from the statement that income taxes cause labour saving. And it matters that one specifies under what conditions income taxes may act as an incentive to do so because an outline of those conditions makes it possible to suggest other policies then simply cutting personal income taxes.

  • Surely tax competition between jurisdictions is one of those unpleasant “facts of life” (like armed forces) that we are going to need to live with, at least until we have one world federal government, and even then, won’t there still be autonomous provinces with at least some rights to set their own rates of tax ?.

    I like to dream of a world in which everyone was vegetarian, no one a “shopaholic”, and where no couple had more than two children. Travis will dream instead (it seems) of a world in which everyone loves paying income tax, and governments agree to harmonize their tax rates. I’m not knocking anyone’s idealism though .. that’s what it takes to keep us going from day to day.

    And for the record, I’m not agitating for income tax cuts out of the blue, or to drastically shrink the public sector. I’m simply looking forward to being able to use consumption / resource extraction taxes to slow down unsustainable rates of ecological depletion, depletion that threatens the entire human economy, BECAUSE it threatens the “wild” ecosystems on which the human economy ultimately depends. It has always seemed to me that some reduction in income taxes is a necessary corollary of doing so, but by all means, lets pursue the primary goal of arresting unsustainable depletion in as distributionally progressive a manner as possible.

    If we can discourage frivolous and wasteful consumption in a manner that is reasonably efficient, allocatively, and if we can do so without cutting income taxes, I wouldn’t be against that.

  • But Michael shifting to consumption taxes will not alter the problem of tax arbitrage, nor tax avoidance. Here too you will either have to think of some minimal international agreement on consumption taxes or be faced with the same problem of arbitrage.

    So yes to change the world we must really change the world. But think how fun it would be to exclaim “international consumption tax accords are just a fact of life…human existence…like war and famine…just the facts.” It is not the “is” we are arguing over here so much as the “ought”.

Leave a Reply

Your email address will not be published. Required fields are marked *