The Unplumbed Revenue Potential of Land

by Mason Gaffney
Professor of Economics, University of Caifornia, Riverside
m.gaffney@dslextreme.com
Understanding Economics

"You see, but you do not observe." — Holmes

The Local Problem: How Assessments Get So Wrong

NOTE: This article appeared in Georgist Journal Nos. 101-103. For a fully documented and annotated version, see Here.

The revenue potential of land is greater than anyone thinks. It should go without saying, but these days does not, that the purpose of raising more land revenues is not to fatten vexatious bureaucrats, but to replace vexatious taxes, to provide needed public infrastructure and services (including a reasonable national defense), to pay off public debts, and to fund social dividends (including existing social dividends like Social Security). This is a progress report on a study that identifies and unclosets elements of enhanced revenue potential by using truer and more comprehensive measures of rent and land values. There are at least fifteen elements of land's taxable capacity that previous researchers have either trivialized, or overlooked entirely. First, we will consider corrections for the downward bias in standard data. The subsequent installments will broaden the concepts of land and its rent, and show how exempting production, trade and capital uncaps potential tax rates. Standard data sources neglect and understate real estate rents and values. These standard sources are both local — assessed valuations used for property taxation, and national — as reported by various national agencies, most of whom use IRS data on reported rents.

I will only enumerate, not elaborate much on the many reasons assessed values usually fall short of the market. Scanning the bullets below, however, gives a clue as to how landowner pressure has subverted the property tax over the years.

The National Problem: Internal Revenue Data

Many economists rely on data generated by the IRS, taken from tax returns, to tell them the sources of income in the US. This is an exercise in crediting bad data. The standard tax procedure of landlords is to deduct alleged "depreciation" from their net operating rents ("cash flow") to arrive at taxable rents. They accelerate depreciation enough, usually, to report little or no taxable rent. This is what the IRS then aggregates and reports as the sum of all rents. To accept such fiction as fact is inexcusable, but economists do it anyway. Their credulity lends their authority to the IRS, while the IRS "official" status helps legitimize the economists — mutual validation of mutual error, the curse of science.

When owner A has exhausted his tax "basis" by overdepreciating, he sells to B for a price well above the remaining basis. B then depreciates the same building all over again, then sells to C, and so on — each building is tax-depreciated several times during its economic life. In any given year, most income properties in the U.S.A. are being tax-depreciated, even though most have already been depreciated at least once.

In addition, all owners after the original builder are in a position to depreciate some of the land value, as well. This is because the owners control the "allocation of basis" between depreciable building and non-depreciable land. The IRS has no defense against secondary owners who overallocate value to the depreciable building. Congress has never authorized the IRS to develop any in-house capacity to value land. The most the agency does, if it will not accept the word of the tax filer, is to look at allocations used by local assessors. These parties, in turn (with a few notable exceptions), underassess land relative to buildings, by using the "land-residual" method. This is partly to accommodate their local constituents — assessors are locally elected or appointed, and do not report to the IRS. A little math will tell you that to depreciate land just once is to achieve perpetual tax exemption. To depreciate it again and again is a continuing subsidy for holding land.

When A sells to B there is a large excess of the sales price over the remaining or "undepreciated" basis. This excess is, to be sure, taxable income. However, Congress has defined this kind of income as a "capital gain." Most rents, therefore, show up as capital gains. These, in turn, are subject to lower tax rates, deferral of tax, forgiveness at time of death and constant political pressure to lower rates to zero. These are known to every lawyer, accountant and Congressman, but apparently not to most economists, who lazily report from "official" data that rents are a low fraction of national income.

In addition, the IRS reports nothing at all for the imputed income of owner-occupied lands, because this kind of non-cash income is not taxable. Todd Sinai and Joseph Gyourko of the Wharton School reported aggregate owner-occupied "house" values in the U.S. in 1999 were $11.1 trillion. The annual rental value of that, figuring at 5%, would be roughly half a trillion dollars a year — quite a chunk to omit from the rental portion of national income. We also know that the prices of lands for both housing and recreation have risen sharply since 1999, perhaps by 50% or so, so that $11.1 trillion may be over $16 trillion now. That means that the imputed rent income is 50% higher than half a trillion (more like, in other words, $750 billion per year), and also that the net worth of the owners has risen by about $5.6 trillion. Such silent gains are also a form of income from land. To all that, many economists remain blind, dumb, and curiously incurious.

Sinai and Gyourko's treatment is superior to what one usually sees, with some effort made to treat land separately. However, even they, like others, write of the imputed income of owner-occupied "housing," exclusively. That is doubly misleading. First, it emphasizes the building. That is wrong because the income properly imputable to the house per se is much less than its rent equivalent. The house requires constant expenses for upkeep, heating, maintenance and repairs, cleaning, painting, etc. The house also depreciates, physically. Those expenses and the depreciation must be deducted from the rental equivalent to get the net income.

The land does not depreciate physically, and so its rental equivalent is its net current income. Usually, it appreciates in value, and that annual increment is also a current income. So the "imputed income of owner-occupied housing" is mostly attributable to the land — but no one is saying so.

Second, the standard characterization of "house values" misleads by omitting vast lands beyond the narrowly defined "house" lot, which includes the land under the building and a little yard or curtilage. What about other lands held for the owners' personal enjoyment? No agency collects data on such lands and their values, but common observation tells us they are vast and valuable, and dominate values in many "rural" counties.

Another Lode of Error: The "NIPA" accounts

The standard source of data on GNP and its components is the National Income and Product Account (NIPA), kept and published regularly by the U.S. Department of Commerce. When it comes to rent, NIPA depends on the IRS figures, which thus are passed along to all students of economics as the "official" accounting. We have just seen how far from reality these data are. NIPA is worse, in a way, because it explicitly excludes "capital gains" from National Income. That is, first the IRS converts rents into capital gains, and then NIPA banishes capital gains from GNP, National Income, and National Product. "Capital gains" is an artificial term, that includes all gains realized from the sale of what Congress defines at any time as "capital assets" — which include land and improvements, housing, common stock, growing timber, breeding herds (including race, show and riding horses), mineral and hydrocarbon reserves in the ground, and several other favorite holdings of the rich and well-connected. As we saw above, most commercial rents show up as capital gains, so that NIPA does not report them at all. Then along come highly visible economists, like Paul Samuelson, Robert Solow, Theodore Schultz, Edwin Mills and Jan Pen, to look up this datum, and declare that land rents, at no more than 5% of national income, cannot possibly support modern governments. This is unfortunate, and quite misleading.

NIPA does, however, make a gesture at including the imputed value of owner-occupied housing. Whether they do it right is a question on my agenda.

Other Prestigious Sources of Error

The Federal Reserve Board is ensnared in the same intellectual webs as the other agencies, so its nominal independence is wasted. Michael Hudson has dissected FRB methods, which resulted in reporting rents of income property far below reality. The reductio ad absurdum arrived when its clerks, evidently plodding "on automatic," duly reported that the rents of all the income property in the USA are negative. Someone in authority finally noticed, was embarrassed, and discontinued the series. Many economists treat numbers from the National Bureau of Economic Research (NBER) as iconic. The press routinely cites their datings of US recessions and recoveries as "official." Many writers cite Raymond Goldsmith's estimates of United States land values, dating from 1955 and 1962, as "authoritative," because they carry the NBER imprimatur. Yet they do not bear examination, even for their times. They were generated as incidents to other work in an offhand and indefensible way.

It is not easy to retrace Goldsmith's steps; one must track interlocking footnotes from several sources. At the end of the trail, however, he simply takes residential land value as 13 percent of real estate value. The basis of this allocation is the share of land in the cost of houses insured by the Federal Housing Authority, which was about 20 percent. (He does not explain why he cut this down to 13 percent.) Goldsmith applies this basis to nonresidential real estate as well. As for corporate-held lands, he enters them at book value — an attitude that opened the door to an epidemic of corporate raiding. Goldsmith also seems to omit vacant lots and unsubdivided land.

These methods are not worthy of the faith with which several economists cite the results. FHA-insured houses are not typical. They tend to be new and on cheap land. Those not new are not very old — in 1967 the median age of insured existing homes was thirteen years. To apply such data to a typical American city, most of whose dwelling units in 1965 antedated 1920, was outlandish then, and even more outlandish today.

FHA clientele is lower middle class, which means the land share is low, land being both a consumer luxury and a rich man's hedge. Land share rises sharply with overall value. The high land share in enclaves of wealth such as Beverly Hills, Greenwich, Belvedere, Santa Fe Springs, Palm Beach or Kenilworth is missing from FHA data.

The FHA is most active at the expanding fringe of cities. A basic fact of urban land economics is that the land share rises toward the center. In Manhattan, for example, the share of assessed land value has always been higher than in the other boroughs.

Applying a land fraction derived from residential data to commerce and industry is not believable. The land share is highest in retailing, the more so now that retailing entails vast parking areas. Filling stations and and drive-ins of all kinds entail vast aprons for small buildings with short lives. Some retailers, such as auto dealerships and lumber yards, store their inventories outdoors. Many wholesalers and industries do the same: tank farms, railroad yards, utility easements, industrial reserves, dumps, salt beds, terminals, heaps of coal and salt and sulfur, and so on and on. In downtown Milwaukee, half the assessed value is land. In Manhattan, it is instructive to consider the Empire State Building. If ever a structure overdeveloped a site, this should be it. Yet in two transactions since 1950 the site was valued at one-third the total. One may infer what this implies of the whole island.

Anyone active in real estate would have caught Goldsmith's error. Yet it passed muster with the NBER, his publisher the Princeton University Press, and several learned academic reviewers. This is not a measure of their general incompetence, but of the extent to which academicians have walled themselves off from anything bearing on the realities of land values and rents. Goldsmith treated land carelessly, as a trivial side-issue, and his finding was ignored by everyone except those who needed to invoke an authority to trivialize land value.

Another Goldsmith error is to exclude subsoil assets. In cities overlying oil pools, like Huntington Beach, that would make a big difference. In most cities that may not matter, but is symptomatic of how insouciantly Goldsmith handled this whole matter of land values.

Ernest Kurnow's Work Under Lincoln and Moley

Ernest Kurnow low-balled land and rent values in a chapter in Theory and Measurement of Rent by Keiper, Kurnow, Clark and Segal, 1961. In an introduction, the authors thank the Lincoln Foundation for financing their work, and go on to thank David Lincoln and Raymond Moley personally for intellectual guidance. Then, extraordinarily, they omit the standard disclaimer which absolves their advisors and takes full responsibility for their own work. This is a unique omission. Res ipsa loquitur: David Lincoln is speaking. That helps explain why researchers seeking full estimates of land values seek in vain at the Lincoln Institute of Land Policy.

Kurnow's basic source is tax assessments. He accepts their allocation of value between land and buildings. He admits that errors are possible, but dismisses them because "in all likelihood there is a tendency for such errors to cancel each other." We have seen how wrong and biased that is. He does not even correct for the assessment bias shown by sales-assessment ratios of Manvel's Census of Governments, nor for the greater degree of underassessment revealed by mapping of land values. He does not consider any of the 18 bulleted points shown above.

In short, the Land Fraction of Real Estate Value is much higher than standard modern sources show. One of many indications is that on most assessment rolls the value of old "junker" buildings, on the eve of demolition, is listed as higher than the land under them. It should be obvious that the old junker has no residual value: that is why it is being junked. Real estate people recognize this concept instantly. It is not obvious to everyone, everywhere, which helps keep it concealed, and provokes a lot of nostalgic resistance. People who make a virtue of recycling old cans and papers can be oblivious to the much higher social value of recycling old urban sites. Many of these old "junkers" even appear sound and valuable, as in enclaves of high values like Winnetka, Illinois, or Beverly Hills, California, but suffer from "locational obsolescence," which is the key concept. That means the growing value of the underlying site for recycling has cannibalized the residual building value.

Most modern economists who look into these matters rely upon the standard sources I've listed here, mindless (or perhaps even glad) of their downward biases. Young students are intimidated and awed, or at least impressed and convinced, by the "official-looking" auspices of the standard sources.

Rents Tappable by Variable Charges

The term "single tax" has been unfortunate in helping to perpetuate a narrow fixation on property taxes; as a result, we tend to underestimate the revenue potential from rents. Many lands and resources that yield rents are not observed or measured in traditional real estate markets. There is a new realization that "taxes on rent" are much broader than the traditional land value tax.

As esteemed a Georgist as William Vickrey often pronounced the prime virtue of land value taxes to be that they are a lump sum, invariant with production or sales. He thus identified them solely as property taxes, and not any variable charge like a severance tax on withdrawing water or oil, a parking fee, a gas tax, or a bridge toll (though he favored all of these, for what he saw as other reasons). He did not see the corporate income tax (which he opposed) as being in part a rent tax. It is a clichι of economics texts to class land taxes together with poll taxes as having the peculiar virtue of not being based on any variable input or output. In this mindset, there are no differences worth mentioning between poll taxes and land taxes — an instance of tunnel-vision that would be surprising in any discipline except, alas, modern economics. Netzer (1973) would substitute "a family of user charges" for taxes on buildings. So strong is the "single-tax" stereotype, though, that not even Netzer (1998) thinks to include user charges as part of land revenues. Then there are mineral revenues from severance taxes and/or royalties, plus income taxes. These are already so great that some polities get much, or even most of their revenues therefrom. And yet the confining "single tax" tradition is so strong that Netzer does not include mineral revenues among land-based taxes — not even in the rents tapped by oil-rich Norway and other North Sea nations. It is a major omission. In one year the mere increase in the value of Norway's undersea reserves exceeded its entire national income.

Variable charges, such as those on road crowding, water withdrawals from surface and underground sources, minerals extraction, air and water pollution, spectrum use, fish catches, billboards, etc., are major additions to land revenues. California, a major oil-producing state, does not even have a severance tax, not even a token. In the fiscal crisis of 2003, with 136 or so candidates running for Governor, only one (Arianna Huffington) even mentioned it, so total is the mental blackout in the state.

If we seek to implement a program of securing the universal right to natural opportunities via the public capture of land rents, then products that cause damage, anti-social behavior and inflated demand for publicly-subsidized medical care may reasonably be taxed. Some examples:

Capturing Rent Via Income Taxation

The income tax base includes income from land. For this we have to thank a few Georgist Congressmen of 1894 who got land included in the base of the income tax which Congress enacted then. In Pollock v. Farmers' Loan and Trust Co., 157 U.S. 429 (1894), the Supreme Court threw out the whole law for that specific reason, leading to the 16th Amendment of 1913 which was necessary, basically, to let land income be included in the base.

Corporate income was successfully taxed from 1909, before the 16th Amendment, as an excise tax on the privilege of doing business as a corporation. "The excise tax used net income as a measure of the privilege of corporate business practice." [Bernard Herber, Modern Public Finance, p. 190.] The legalistic circumlocution suggests how creative lawyers can implement what Congress really wants. Someday another text might read "The excise tax used land value as a measure of the privilege of holding title to natural resources."

But that may not even be necessary now. State legislatures, like Congress, have nearly complete control and discretion over what kinds of income to include or exclude from the income tax base. They have abandoned most of their discretion by piggybacking on Federal laws, but they have not abandoned all of it, and they could take it all back.

The income tax can be converted into a tax on land income in two steps. The first one is surpassingly simple: exempt wage and salary income from the tax. One could tiptoe up on this by raising the earned income exemption, the standard deduction, personal exemptions, etc. Workers paying the social security tax should be allowed to deduct it from taxable income. Raise the rates on what remains of the income tax base, which would now be mostly property income. If that seems shocking or radical, recall that from 1913-41 (before withholding, and the explosion of the FICA deduction) most wage and salary income was in fact exempt. What is really shocking and radical is the massive shift of tax burden off of property income and onto wage and salary income, a shift that has perverted the whole notion of income taxation as originally adopted in 1913.

The second step is to remove capital income from the base. This is harder to understand, but easier to accomplish because it has already been done in part. The present tax law includes several devices designed to lower or effectively eliminate any tax on the income from capital. Basically, this is done by letting investors write off what they invest at or near the time they invest it. The investment tax credit (ITC) even goes farther and lets them write off more than they invest.

"Expensing" of certain capital investments means writing them off 100% in the year made. Accelerated depreciation is a substantial move in the same direction. Even straight-line depreciation is really accelerated compared to the true depreciation paths of durable capital, especially when coupled with the use of tax lives which are much shorter than economic lives of durable capital items.

None of those devices apply to land, however, because land is not depreciable. That is again thanks to generations of Georgists, starting with those in the Progressive movement when the income tax was shaped. Who else would keep officials conscious that land is different? Standard-brand academic economists keep pushing the notion that land is just a form of capital.

To convert the tax fully to land, then, we need only complete step two by allowing universal expensing of all new investments. Voila!

At the same time we must plug many loopholes designed especially for land income. One of these is depreciating land, even though land does not wear out. This is illegal, strictly speaking, but it is often winked at in practice when old buildings are depreciated from their purchase price by new buyers.

Many will object that the income tax only hits realized income from land, and exempts the holder who neglects or underutilizes land. True enough — but consider the behavior of private landlords and tenants. They often prefer arrangements that share risks and returns, like the income tax, instead of fixed cash rents that resemble the property tax. The cases are not perfectly analogous in all particulars, but suggestive.

It seems clear that, should a legislature wish to go further in this good direction, it could define "land income" as a fixed proportion of land value, regardless of use. Plenty of economists would come forth to testify that that is a reasonable definition.

Substituting Taxes for Subsidies to Promote Conservation

Here is a high potential to turn "Negabucks into Megabucks" for the treasury. For generations, we have subsidized landowners to withdraw water. The benefits of the subsidy have gone roughly in proportion to the area of irrigable land owned. As a result, water is maldistributed, underpriced and wasted. Today, for a change, there is support (at least intellectually) for a groundwater extraction charge, purely as a conservation and efficiency measure, and to obviate megabuck "rescue" projects. However, if we can wrench our mindsets away from the crazy tradition of subsidizing waste and maldistribution, there is also great revenue potential in water. In an arid land, water is life. Some, perhaps much of the land rent now imputing to fee simple lands can be transferred to the holders of water, simply by raising its price.

Why should we want to transfer the burden to the holders of water? Because a state's water belongs to its people. A license to withdraw the people's water is not real property (and thus not sheltered by Prop 13). The State can serve free market efficiency and raise revenue in one stroke by putting a charge on water withdrawals. Such a charge would expedite the powerful current movement to market water.

An economic charge should of course be geared to the economic value (locational, mainly) of waters. Groundwater has been mentioned. Surface water could bear higher charges because it is already at the surface with no pumping. This charge might be called a "tax," or a rental for state property, as legalism and politics may require. The charge should cover not just active withdrawals, but "dog-in-the-manger" licenses to block withdrawals by others. Value-data to help set a proper charge would come from the proposed free market in tradable water licenses.

Unearned Increments as Current Rents

There is a swelling of "capital" gains (mostly land gains, actually) as a component of income. In this case there is no corresponding realization among economists or the public that capital gains on land are eminently taxable. On the contrary, as gains grow so do the wealth and political power of the movement to untax them. So much greater, then, is the need for objective economists to establish the taxability of capital gains. Unrealized gains can be taxed as they accrue, without disincentive effects or administrative nightmares, and economists need to estimate the new revenue potential that now largely escapes taxation.

Capital gains as a revenue source can be quite unstable. California's current (2003) fiscal bind illustrates the problem. This should not be taken to be a drawback of the present proposal, however, for the proposal here differs from the current income tax on capital gains in several ways.

My proposed tax is focused on unearned increments to land values. Current income taxes include gains from a variety of other sources, like building up a new business. During the dot.com boom, it was this last element that was most unstable.

My proposal is to tax land-value gains as they accrue, rather than upon sale. During a land boom and bust, land taxes are a strong stabilizing factor.

My proposed tax excludes gains on common stocks.

Variant kinds of land resources, hitherto neglected or not classed, or only recently classed with land, show great revenue potential. Some examples are the radio spectrum; telecom relay sites; slots in the geosynchronous orbit; fishing quotas; quotas of all sorts on production and marketing; pollution permits; licenses to withdraw water; power drops; parking spaces; highway access; mooring spaces; etc. Many in the Green Movement see the double efficacy of Pigovian taxes to curtail overuse and pollution of common airs and waters, while also raising revenue (Costanza, et al.). (Many academicians, sadly, are dragging their feet and making themselves part of the problem by bickering over whether this is possible.)

Uncapping the Tax Rate on Land

The standard reasons for avoiding high tax rates (and, rather, spreading low rates of taxation around to many sources) do not apply to land. Ownership of wealth generally, and land and capital gains particularly, are highly concentrated. They are much more concentrated than incomes from productive labor, and increasingly so. Thus, taxes based on land rents and values are progressive in their impact and incidence, at the same time they are pro-incentive in their allocative effects. This combination of virtues is unique. It belies the clichι that governments must always choose between equity and efficiency in taxation. It makes it possible to raise tax rates to high levels without either stifling good incentives or embracing regressivity. This greatly enhances the revenue potential of such taxes.

The Unseen Reservoir of High Internal Valuations and Holdout Prices

Observed land markets understate the value of land to most landowners. These owners' internal valuations are above the observed market: that is why they do not offer to sell. In most land markets, annual turnover is 5% or less. Assessors take that sample to estimate the value of the whole. The other 95% of landowners in effect "sell" or "rent" to themselves each year. How accurately does that 5% sample the entire invisible "market" for land? Many owners routinely declare "Get away from my door; I will not sell for any price."

Modern environmental economics has spawned the discipline of "contingent valuation" to appraise damages to resources that seldom pass through markets. It turns out there is a major difference between WTP values (what will you pay for cleaner air) and WTA values (what I must pay you for permission to pollute your clean air). WTA >> WTP. Where there are market dealings to observe, they are based on WTP values, so the observed market conceals WTA values, which are much higher than the active, visible "market." The "willing seller" concept is mostly fictional: it is the "motivated seller" who makes the market — the observed market, that is. Most sales are in some way "forced." Other owners hold out for much higher prices.

Status-quo theory is shaken to the roots by survey findings that WTA >> WTP. If we acknowledge the common birthright to a clean environment, then you can't pollute anyone's air or water, because the victims own it. They can be as unreasonable as any great landlord. This explains why theorists are so busily trying to plug the dike. It was 1974 when a survey first showed WTA >> WTP, "in contradiction to received theory." This sent dozens of professors and think-tankers scurrying to torture data and logic until they confessed otherwise, to save Coase and Stigler. Mitchell and Carson, for example, slog through a long literature survey, apparently impartially, but in the end find ways to stick with WTP after all (1989, pp. 37-38). They have succeeded in keeping the mass of economists in denial on the matter, so economists don't even see its implications. The meaning for tax policy is that there is scope for substantially raising tax rates on land without flooding the market with distress sellers. That will disappoint those (including myself) who see land taxes as a means to cheapen land for new buyers. That goal will take high tax rates; but en route to the goal (and also afterward) we can raise great revenues, which is the present point.

The flipside of high internal valuations by owners is that roughly one third of American families are renters. Their internal valuations of what they rent are obviously lower than the market value of these or comparable quarters.

ATCOR (All Taxes Come Out of Rents)

When we lower taxes, the revenue base is not lost, but shifted to land rents and values, which can then yield more taxes. This is most obvious with taxes on buildings. When we exempt buildings, and raise tax rates on the land under them, we are still taxing the same real estate; we are just taxing it in a different way. We will show that this "different way" actually raises the revenue capacity of real estate by a large factor. There is much recent historical experience with exempting buildings from the property tax, in whole or part. It has shown that builders offer more for land, and sellers demand more, when the new buildings are to be untaxed. The effect on revenue is the same as taxing prospective new buildings before they are even built, even though the new buildings are not to be taxed at all.

Land value is what the bare land would sell for. It is specifically and immediately most sensitive to taxes on new buildings, and on land sales, as well as to new and more stringent building code requirements or zoning that often discriminate against new buildings. Where new buildings are "coded" more severely than old, it enhances the value of the old land/building packages. This premium should be considered part of land value, and taxable as such.

We have numerous historical experiences with exempting buildings leading to land booms: New York City 1922-33, Western Canada, Hong Kong, Taiwan, Australia, South Africa, San Francisco after the fire, Chicago after the fire, California Irrigation Districts, Cleveland 1903-20, Toledo, Detroit, Portland, Seattle, Houston, San Diego.

Familiar Micro Cases

The general principle that tax cuts shift to higher rents is, in many ways, like the forest: too ubiquitous for most to see clearly. But here are a few of the trees:

The Resource Curse Effect

Economists and historians have noticed that nations and regions that are rich in natural resources to export often lag in manufacturing. This is often now called "The Dutch Disease," although obviously they did not catch it until modern times, with the oil and natural gas booms. These prized exportable items raised the value of the Guilder, making Dutch manufactures cost foreigners more, and letting Dutch consumers import competing foreign products. Canada exports lumber and energy products to the same effect; so does Alaska, which also collects great federal largesse, military and porkbarrel. Canada taps into resource revenues to lower national taxes; Alaska, to lower other State taxes and distribute social dividend to each resident. Thus, resource rents help raise other land values.

Utility-Rate Effect

Lower rates mean higher land values. During the Progressive Era, rapid growth of cities called for providing costly utilities and transit on a new and massive scale. Many big-city mayors, some directly instructed by Tom Johnson of Cleveland, saw that providing these services raised land values, which could be taxed to pay for them. Private franchisees saw they could profit by squeezing monopoly profits from the franchises. It became a running battle.

The economics profession lagged in responding. A number of professors were removed from leading universities after writing or speaking too openly against the franchises (Tom Johnson hired one of them, Edward Bemis, to advise him on rate regulation). Rich franchisees, after all, might help endow universities as Chicago traction magnate Charles Yerkes did with his observatory. But in 1938 Professor Harold Hotelling of Columbia drew the point sharply in a leading article in the obscurely statistical journal Econometrica. He was followed over time by a school of thinkers who favor "Marginal-cost pricing," which often means lowering user rates on mass transit and utilities, making up the deficits by taxing the benefited lands. Theorist Abba Lerner even tried to squeeze all of economics into what he called "The Rule": set price equal to marginal cost. More recently Martin Feldstein, William Vickrey, Joseph Stiglitz, Richard Arnott and others have popularized what they call "The Henry George Theorem" wherein lower utility rates lead to higher land rents.

The Logic of ATCOR

The thesis that all taxes are shifted to landowners follows logically from two premises. One, after-tax interest rates are determined by world markets. The local supply of capital is perfectly elastic at a fixed, after-tax rate. Two, the wages of labor has been reduced to so low a level that it cannot bear any more tax burden. Anyone may test the premises by observation. If there are unemployed workers, then the supply of "work" (as opposed to "labor," defined as so many warm bodies) is highly elastic. When we find work for the unemployed and underemployed, labor gains without costing land or capital anything at all. Even better, in fact, labor gains while benefiting other taxpayers, because of lower dole costs, lower crime costs, etc. The enhanced psychic benefit of universal job security is also worth a lot (although not in direct money). In the era when Keynesianism was in flower, many alleged that the social cost of putting the unemployed to work is zero.

Nowadays, of course, Keynes is out of style with the dominant anti-labor schools. Unemployment is considered simply as leisure — a voluntary choice, a matter of personal taste. And yet, Chicagoans like Gary Becker freely postulate elastic labor supplies when they blame unemployment, as Becker routinely does, on minimum wage laws. It's not clear whether Becker sees the parallel, but that is a Keynesian assumption.

It is likely that real wage rates would rise, as more-efficient land use increased demand for labor and lowered product prices. Compact settlement would create new rents via the synergies that are not aborted by scatter. This was the theme of Progress and Poverty, and the primary goal of Henry George's reforms. True, that was before we had heavy payroll and income taxes on labor. In real terms, though, the outcome is the same: it is likely that the abolition of such taxes would let after-tax wage rates rise, even while before-tax wage rates remain the same, or fall. To the extent that this process diminished, if it did, the overall public-revenue potential of land, few would call it a calamity.

Capital Supply is Elastic

Most economists assume this, emphasizing world markets, rapid transfers, arbitrage. However (and in addition), even in small closed economies, there is underemployed capital, just like labor. This is because the return is held down by taxation. So it goes into untaxed consumer goods, and tax-exempt forms of capital, like housing, foundations, government works or personal property. When all uses of capital are untaxed, these forms would be placed on equal footing with higher-yield opportunities. From this would spring a large supply — voila! elasticity in the supply of capital. George recognized this, although he had his own way of expressing it. He did not regard consumer capital as being "really" capital (as it was not actively being used in production), but he did observe people living on it while they produced other capital. During World War II we experienced a vivid example of how people can draw down consumer capital to meet an emergency need.

Logic and experience both overwhelmingly support the idea of ATCOR. To summarize: the revenue capacity of land, when it is substituted for other tax bases, is comparable to current revenues. Owing to efficiency effects, and renewal effects, it may well be higher. The major reservation is that the supply of labor is not totally elastic, so some of the revenue gains may be "lost" in higher wage rates, but on the whole higher wage rates are socially desirable, and serve to lower many public costs as for welfare, policing and jailing, aggressive military spending, make-work projects, etc.

Multiplier Effect of Taxing Absentee Owners

Transferring rents from them to our fisc, and spending the proceeds locally, improves the state economic base and balance of payments. It is alleged that we must avoid taxing absentees, because they will remove their capital from our state, but they cannot remove most land. The only way they can remove oil and gas is by producing them. The present owners of most of our oil and gas became so by acquiring it from existing local owners and producers, so it is hard to argue they ever did bring capital into the state. It is easy to argue, however, that a democratic sovereign state reports to and is responsible to the resident electors, not absentee owners. It is easy to argue that the quality of life is worsened when absentee owners displace local owners and turn local people into tenants. There is no social value in encouraging absentees.

A high percentage of real property is owned from out of state and even out of the country. The percentage is much higher than we may think. It is not just Japanese banks and the Arabs in Beverly Hills. It is corporate-held property which comprises almost half the real estate tax base. If we assume that California's share of the stockholders equals its share of the national population, then ninety per cent of this property is absentee-owned; the percentage may be higher because many of these are multinational corporations with multinational owners.

There is a curious silence on the matter. Some critics of capping the property tax rate talk about "business" securing the lion's share of benefits. No one seems to have seized on the fact that half the taxable property in California is owned by people not voting in the state, and not spending their incomes in the state. Here is one instance where localism (which can be ugly, as we know) may be harnessed to help create a more healthy society. The purpose of democracy is to represent the electorate, not the absentee who stands between the resident and the resources of his homeland.

California's legislative analyst, William Hamm, estimated in 1978 that over fifty per cent of the value of taxable property in California was absentee-owned. This is such a bold, bare, and enormous fact it is hard to believe that Californians could be misled into resisting the urge to levy taxes on all this foreign wealth. They may be put off by the argument that they need to attract outside capital, but that carries no weight when considering the large percentage of this property which is land value. Some half of any reduction in California property taxes leaks to out-of-state owners. Nor is this the only leakage. Net federal income tax payments have risen because sales and nuisance taxes raised to replace lost property taxes are not deductible. Sales of local general obligation bonds have stopped and will stay stopped. Revenue bonds are sold instead, with higher interest rates. Fire insurance rates must rise. And private spending substituted for public spending will have a higher propensity to import. Public spending goes for policemen, firemen, teachers, local contractors, and so on.

This substantial leakage of economic base results in multiple declines in state income. One drastic example of this is offshore oil and gas, which is outside state sovereignty and escapes all state and local taxation. One result is unbalanced state hostility to offshore leasing, for the locals suffer the degradation without sharing the gains. Some provision for state sharing in offshore revenues seems indicated.

The Picture So Far

In this article and our two prior installments, we have discussed fourteen new elements of land's taxable capacity. Previous estimates of rent and land values have been narrowly limited to a fraction of the whole, thus giving an entirely false impression that the tax capacity is similarly narrow. We are adding the following elements to the traditional narrow "single tax" base:
  1. Correcting omissions and understatements in standard data sources
  2. Updating ancient sources that use obsolete low values
  3. Raising the Land Fraction of Real Estate Values (LFREV)
  4. Adding rents that are best taxed by use of variable excises
  5. Adding rents taxable by income taxes
  6. Substituting taxes for subsidies to foster conservation
  7. Adding current unearned increments as part of ongoing rent
  8. Adding previously invisible and undervalued resources to the tax base
  9. Adding lands held under variant forms of tenure
  10. Adding rents that are now dissipated, but need not be
  11. Noting the feasibility of much higher tax rates on a base that is both non-erosive, and concentrated in ownership
  12. Noting the great mass of holdout prices (WTA values) exceed visible market prices (WTP values) by a large factor
  13. Adding the revenue from most existing taxes to the potential land tax base, on the ATCOR principle
  14. Multiplier effect of taxing absentee landowners
Any one of those Fourteen Elements indicates a significantly higher land tax base than economists commonly perceive today. Taken together, they are overwhelming, and cast an entirely new light on this subject.

One Final Rent-Raising Factor: Mortgage Interest as Land Rent

Here is one further supplement to the land rent tax base, which I am not counting among the basic fourteen because it involves novel thinking, and is fraught with controversy, which might divert us too much from the main chance.

One kind of paper is systematically recorded at the county level: mortgages, or deeds of trust. It is administratively feasible to put these into the property tax base, as Professor Don Hagman kept urging. But is it desirable? A tax on mortgages would be mostly shifted to borrowers in the form of higher interest rates, the supply of mortgage funds being highly elastic. Thus, to tax mortgages is indirectly to tax real estate.

It is widely assumed that cheap long term credit is essential to let most people buy real estate. Unfortunately that reasoning overlooks the nature of land values, which makes it circular. The main effect of long term loans has been to inflate land prices, creating the very problem it offsets. It is a treadmill effect, like keeping up with the Jones's.

It must be conceded that holders of existing mortgages would suffer. But someone suffers with any change of tax or other public policy; there are always winners and losers. It is a risk all investors take knowingly. Phasing-in is possible, and it should be remembered that in a Georgist tax shift, most holders of mortgages would be relieved of some or all of the income-tax burden they currently endure. (Another benefit of including mortgages in the property tax base is to counter the argument that the property tax discriminates against equity holders of real estate. Many have questioned the equity of focusing taxes on the person with 5% equity in a parcel, while exempting his bank.)

Would new lending be discouraged? Yes, at the margins. The most sensitive margin is one which most people would not perceive at first, that is the margin of durability or longevity. The more deferred the benefit of an investment, the more interest-sensitive is its present value. But, is that bad? We are conditioned to answer "yes," but as an economist, I doubt it. The financial system will adapt by basing loans less on land collateral, and more on buildings, inventories, accounts receivable, crops, personal reputation, and appraisal of specific projects. This is more labor-intensive banking, and less capital-intensive. Untaxing labor, as proposed herein, makes this more feasible. On balance this will help stabilize the financial system, whose worst fiascos, like the South Sea Bubble of 1720, the world banking collapse of 1932, the American Savings-and-Loan debacle of 1987-91, the Japanese collapse after 1992, and so on have resulted from speculative loans on land. It is time to revive the old "commercial loan" theories of banking, with their emphasis on liquidity and quality of credit, achieved mainly by sticking to self-liquidating short-term commercial loans, and avoiding long and speculative loans secured by real estate. It is a subject too big to open here, but you will find plenty of support in the history and theory of banking for keeping lenders out of mortgages.

Rent as Revenue: Quantity and Quality

I hope that this brief survey has demonstrated land's suitability as a tax base in terms of quantity. It is eminently suitable in terms of quality as well. The macro-economic benefits are deep, wide, high, and temporal. To produce the added goods the owner invests more capital and hires more labor, or sells parcels to laborers wanting to go into business on their own. The newly employed labor earns incomes to buy the newly produced goods and services. Here is supply-side economics coupled with demand-side economics. The conventional left-wing objections to Say's Law do not apply here, because we are untaxing capital at the same time, and raising investment opportunities. We are financing government to provide needed infrastructure to develop new lands, or redevelop brownfield lands, to open new investment opportunities for private capital. The conventional right-wing objection that capital is limited does not apply, either, because we are stimulating saving, stimulating import of capital, and raising turnover of capital. (Turnover raises the ratio of income-creating investing to capital.) It is the macro-economists dream, leveling upwards while balancing supply and demand, saving and investing.


Back to Understanding Economics