- Further review reveals the importance of housing values for Piketty's thesis and suggests some counterintuitive results reached in his work.
- Piketty's recommendation for very high marginal income tax rates would lead to a preoccupation with tax avoidance and his tax on capital idea is unworkable.
- A better solution for the United States would be much lower income taxes, a value added tax and other policy revisions targeted at our real problems.
1. More Math - Piketty's argument that the ratio of capital to national income has increased dramatically in the past several decades relies heavily on his finding that the amount of capital attributable to housing has skyrocketed. His Figures 3.1, 3.2, 4.1 and 4.6 demonstrate this trend for (respectively) Great Britain, France, Germany and the United States. In the case of France, the trend is powerful and it appears that housing wealth has gone from roughly 100% of national income in 1950 to roughly 400% of national income in 2010. Indeed, but for this increase in housing wealth the ratio between capital and national income in France would have remained remarkably steady over the past 60 years. In Great Britain, the trend is also powerful but less extreme, leading to an elevation of housing wealth to roughly 300% of national income in 2010. In the case of Germany, housing wealth seems to have gone from roughly 50% of national income to 200% in that time period. In the United States the trend does not appear to be sharp at all with housing wealth hovering around 200% of national income for the entire period. So what exactly is going on?
First of all, it is important to note that in all four countries there is a high percentage (estimates are - US - 65%, UK- 67%, France - 64% and Germany - 53%) of owner occupancy in the residential housing sector. Because housing is viewed through the window of imputed rent, this means that the increase in housing values implies an increase in imputed rent. For example, if Phil Mause the worker earns $80k per year and buys a house for $200K with an imputed rent of $16K per year, Piketty's model presumptively has Phil Mause the landlord/capitalist charging Phil Mause the worker $16k. If the property goes up in value to $400K, then Phil Mause the landlord/capitalist/rentier is now charging Phil Mause the worker rent in the amount of $32k per year and we can see that returns on capital are eating up a bigger share of annual income. Phil Mause the worker is being squeezed to death by higher rents while Phil Mause the rentier is fat and happy.
Of course, this is very artificial. The real Phil Mause notices no difference except for the wealth effect (he feels richer because he notices his house is worth more on Zillow). So, if the major cause for the increase in the ratio of capital to national income is a dramatic increase in housing prices, the notion that the impact is that "rentiers" are misappropriating a larger portion of the national wealth is largely (at least to the extent that housing is owner occupied) a product of artificial accounting.
A Tale of Four Cities - In the case of France and Germany (and to a lesser extent Great Britain) the Second World War destroyed huge amounts of housing stock and in 1950 rebuilding still had a long way to go (especially in Germany which now includes what used to be East Germany). Indeed, when I traveled to East Berlin in the early 1970s, it looked like it had been bombed very recently and the devastation was almost as bad as what I observed in Detroit 25 years later. So it is not surprising that housing values increased in the period after 1950. In addition, our current ratios may be very much affected by the data coming from large and very expensive cities. Paris and London have notoriously expensive housing and data from these cities looms large in national aggregate data. The table below shows the metropolitan area populations (in millions) for New York, Berlin, London and Paris, the national populations for the four countries and the percentage of each country's population living in each metropolitan area.
|City||Metro Pop.||National Pop.||Percentage In Metro|
We can argue about the proper definition of a metropolitan area and there is room for debate about the precise numbers but the general relationships are clear. In France and especially the UK, a very large percentage of the population lives in the capital city metropolitan area. Indeed, this data and the fact that London is generally considered to have more outrageously expensive housing than Paris would suggest that the UK and not France should have the highest ratio of housing capital to national income. At any rate, Paris and London each play a role which no American or German city can rival. They are the clearly predominant financial, intellectual and tourist destinations hubs as well as being the national political capital of their respective countries. In the United States, we would have to move all government operations in Washington, D.C., and Albany as well as Harvard University into New York City to replicate the role London and Paris play.
So housing in London and Paris is very expensive. One survey suggests that my house (in a nice part of Washington, D.C.) would be worth 20 times its Zillow price estimate if it were in London. As noted above, surveys seem to indicate that London is more expensive than Paris and this leads me to question Piketty's numbers but, at any rate, both cities are very, very expensive and this tends to inflate national totals for France and the UK.
Understanding Urban Real Estate - As one who has invested in urban housing, it has become clear to me that the most important element in the valuation of urban housing is government regulatory policy. First of all, the value of a piece of property depends not so much on the land or the building but, instead, on the development rights. A given piece of land in an historic district for which development is restricted to a strictly conforming two-story single family residence is worth a fraction of the price of the very same piece of land if it is zoned for high rise residential development. Restrictions on development in an area also tend to drive up the prices of existing structures in which development rights have been "locked in" by the construction of a large building.
Of equal importance in the residential housing market is rent control and other regulatory policies controlling landlord/tenant relationships. Indeed, in an extreme rent control situation, the actual fee simple owner of a attractive building in a desirable location may have an asset with a very low value. The real value or "capital" may be held by the tenants who have the legal right to stay in place at rents dramatically below fair market value. I'm not sure how a calculation could be done but the expected discount from market value over the expected tenure of the tenant could be used to generate a discounted present value of the leasehold and, in cities like New York, in many cases it would be considerable. Of course, if we attribute this "capital" to tenants in rent controlled buildings, we would have a very different distribution of wealth than that indicated by the calculations of Piketty.
So, what we really have is a series of government policies - the decision to centralize operations in one city, restrictions on development, and rent control - which create a very expensive housing market (for those not in place under rent control) and create some enormous windfalls for various players. In the case of France and the UK, this seems to be the primary explanation for the build up of "capital" relative to national income. Of course, using this phenomenon to "prove" Piketty's thesis - that capitalism inevitably leads to a growth of capital relative to income - is preposterous. Indeed, the problem he "uncovers" is the result of government policy and regulation and not any inevitable defect in capitalism. The solution is not a wealth tax but rather less restrictions on development, policies designed to decentralize various activities and, perhaps, a relocation of national political capitals.
2. Piketty's Policy Prescriptions - Piketty does not really offer a very well defined solution to the problem he purports to identify. He tentatively suggests "modernizing" the social state, recognizing that "once the public sector grows beyond a certain size, it must contend with serious problems of organization."(p.482) He advocates greater "transparency" with respect to education, merit and wealth but provides no real details on what he would propose nor does he even define very well what he is talking about. Of course, it is always hard to oppose transparency but it is impossible to agree or disagree with Piketty on this point because he is so vague. Piketty also talks about pensions in vague terms, again advocating that greater clarity would be beneficial. I'm not familiar with the French system but I would tend to agree that greater efforts should be made to help people understand exactly what they are likely to receive and what their choices are.
Without being very specific, Piketty advocates progressive taxation and appears at one point to support a tax of 50 or 60 percent on all incomes above $200,000 in the United States and possibly as high as 80 percent on incomes over $500,000 (p. 513) but he appears half-hearted about this and concedes it is not likely to happen. My own experiences with a 70 percent marginal tax rate early in my career are consistent with the expectations that would be generated by reasonable microeconomic analysis. An enormous amount of time and effort is devoted to tax avoidance when rates get above 50%. There also is a tremendous incentive for tax evasion and one begins to see a preference of "cash" businesses and all sorts of barter transactions designed to eliminate that hated thing - ordinary income. The inefficiencies associated with this kind of system swamp any distributional benefits. In the United States, we have "been there, done that" and I hope we don't ever go back.
His big new idea is a "global tax on capital" (pp. 515-39). Although he is somewhat vague concerning rates of taxation, he clearly wants the tax to be progressive. He also recognizes that a great deal of wealth is taxed in the form of real estate and property taxes (p. 517). Given the enormous percentage of capital currently allocated to housing (especially in France) and the fact that there is a considerable amount of capital allocated to other forms of real estate as well, one would think that this would do the trick. But Piketty likes the idea of a capital tax, if only at 1 percent a year, because it would promote international cooperation in monitoring, reporting and ultimately controlling capital flows. One suspects that Piketty may selfishly advocate this measure partly because it would help generate the data that could enable him to produce a sequel to his book. At any rate, the implementation of his proposal implies a level of international cooperation which, if it existed, could be harnessed for so many other worthwhile objectives (arms control, climate change abatement, disease control, etc.) that capital taxes would not even be on the first page of the list.
3. Alternate Policy Proposals - My alternative proposals are premised on several assumptions. I assume that wealthier individuals should bear a greater proportionate share of the burden of supporting national government. I want a simpler tax system which reduces transactional costs. I want to avoid marginal rates that are so high that they induce high income individuals to spend most of their time and effort at tax avoidance and tax evasion. I want a system that puts US industry on a level playing field with foreign competition.
A. Value Added Tax - The federal government would rely to a very great extent on proceeds from a broad based value added tax. Because this tax would be imposed on imports but forgiven on exports, it would put our industries on a level playing field with foreign competition (virtually all of our major trading partners have this tax system).
B. Income Tax - All payroll, social security and Medicare taxes would be abolished. An income tax with deductions only for state and local tax payments, charitable contributions and a mortgage interest tax capped at $100K and phasing out over 20 years would be imposed. The top rate would be 25% on incomes over $2 million. the first $100K in income would be exempt and there would be no marriage penalty. The rate for $100K to $250K for an individual would be 10% with the rates scaling up to 25% for incomes over $2 million. A $25K dependency allowance would be permitted so that a family of four would pay no tax on the first $250K of income. Inheritances would be taxed as ordinary income with averaging permitted. Dividends would be taxed as ordinary income. Short-term capital gains would be taxed as ordinary income and long-term capital gains would get reduced maximum rates based on the length of the holding period (to compensate for inflation).
C. Corporate Tax - A much lower corporate tax rate of 10 percent or 15 percent would be employed.
D. Estate Taxes - Estate taxes would be limited to a tax of 10% on the amount of stepped up cost basis on assets. Assets would be distributed to heirs at a tax cost basis equal to current fair market value so that the heirs would pay reduced capital gains taxes if they sold the assets.
E. Negative Income Tax - A negative income tax of $5,000 would be paid to compensate low income individuals for the impact of the value added tax. This would scale down starting at incomes of $25,000 and would be eliminated at the threshold of $100,000.
F. Gasoline and Diesel Fuel Taxes - A tax starting at a low level and increasing over ten years to roughly $2 a gallon would be imposed on gasoline and diesel fuel but could be reduced at that point in time at which the United States was no longer a net importer of petroleum and petroleum products.
G. Fiscal and Monetary Policy - Due to the concentration of income among the top one percent, expansive fiscal and monetary policy will become necessary because volatility in consumer spending will increase. On the fiscal side, a priority list of infrastructure projects will be developed so that they can be implemented whenever unemployment falls below a certain level. More long term bonds (50 year, 100 year and consols) will be used to finance deficits. Monetary policy will be expansive unless and until inflation becomes a threat.
H. The Housing Problem - In large cities experiencing overly expensive housing, restrictions on development should be eased to expand supply.
I. France and the United Kingdom - Based on Piketty's data, the enormous explosion in real estate prices and rental costs in London and Paris can best be alleviated by permitting more development and relocating national capitals - in the case of the United Kingdom north toward the center of the country (perhaps Manchester) and in the case of France certainly not to Vichy but toward the Bay of Biscay (perhaps Bordeaux). The relocation can easily be financed by the sale of real estate into an ultra-expensive market. Germany, Russia, the United States, Israel and Brazil have all relocated their capitals; it is hard but not impossible.
4. Conclusion In Part 1, I demonstrated that the central premise on which Piketty's thesis is grounded is incorrect. Even if r>g, the ratio between capital and national income will not necessarily increase. Indeed, it appears that capital growth will tend to hit a kind of "terminal velocity" at which level the need to offset depreciation as well as the low return on capital will make it very hard for capital to grow without enormous levels of savings which do not appear to be plausible.
In Parts 2 and 3, I demonstrated that a great deal of the increase in capital between 1950 and 2010 was due to a precipitous increase in the amount of capital designated as housing. For several reasons, this appears to be a more acute problem in France and the United Kingdom than in the United States. To the degree that the increase in value has been experienced in owner occupied housing, it is hard to see this as any kind of problem necessitating corrective measures. To the extent that it has occurred in rental housing, it appears that Piketty has written a 685 page book whose central thesis boils down to the fact that rents are too high in London and Paris. This does not constitute any kind of indictment of capitalism in general. It does, however, suggest other remedial measures which I describe.
The rich should pay a higher percentage of their incomes to support the government than the poor and the middle class. The best way to get to this result is through a broad based, simple tax system that reduces the opportunity for avoidance. Tax rates in excess of 40% lead to an inefficient and damaging preoccupation with tax avoidance and tax evasion and are a ham-handed approach to the problem.
Piketty has contributed important insights and research results but he failed to explore the implications of housing and its outsized impacts on his results. His work should lead to a robust debate on these issues and should encourage further research into both historical and current wealth and income data.