A Cannan Hits the Mark

By Mason Gaffney

Edwin Cannan (1861-1935) is best known for his 1904 edition of The Wealth of Nations, which became a standard.  His other best-known work is a History of Theories of Production and Distribution, 1893.  His book most relevant here is History of Local Rates in England, 1896.  He was a professor at the London School of Economics, 1907-26, although a large inherited fortune let him live and rub elbows at Oxford, which he seemed to prefer.  His later work was less noteworthy.  He criticized both Marshall and J.M. Keynes, but without much impact.

Cannan’s Law

In 1907 Cannan fired off a round at local rating of site values.[1]  It hit home.  First he recited the logic of what today we call the “tragedy of the commons” (it was common coin long before Garrett Hardin).  Then he pointed out that a city taxing only site values to provide free public services would attract too many people and too much capital.[2]  A city is an “open economy,” free to immigration of everything but land, something like an open range or fishery.   Even if all cities tax only site values, cities with more rents per head may support public services at higher levels, and so attract immigrants.  This distorts locational decisions, attracting people to jobs of lesser productivity where they may gain from better public services.  This is “Cannan’s Law.”

There are three bad results from Cannan’s Law.  One is an uneconomical distribution of population, as cities with more rentable lands attract more of mobile labor and capital than they should.  That is not to deny that people are attracted to New York for good economic reasons.  Rather, it is that distributing economic rent freely to all comers attracts people above and beyond the good economic reasons.  Thus, people move to New York to earn high wages, well and good; but in addition they may receive a high quality college education from CCNY, the “poor man’s Harvard,” paid from local property taxes. In the glory days of the Mesabi iron range, children of immigrant Finnish miners there in Hibbing, Minnesota, enjoyed some of the best schooling in the country, paid from local property taxes on iron ore.  In Alaska and Alberta, workers receive high wages to overcome the harsh climate, remote locations, and other disamenities.  That is economically sound, but in addition they get a cash dividend each year from the overflowing oil revenues.  All that tends to draw more people, like flies swarming to fresh pie, than the wages warrant.

A second bad result is what economists call “dissipation of economic rent.”  To make it simple, consider a rich but crowded fishery where another fishing boat added to the crowd will not raise the total catch at all, but simply take fish from other crews who were already there.  Interlopers will keep entering until the average boat and crew just make costs, leaving no net rent for anyone.  This has long been standard economic lore.  As Cannan writes, if a locality uses its rents to benefit all its “inhabitants,” people will flock to the richest places until there is no further gain to immigrants because they have wiped out all the rent.[3]

A third bad result of Cannan’s Law is to lower the incentive of local governments to provide public services that are open to all comers.  It fosters local institutions and attitudes that are harshly hostile to newcomers and outsiders, especially to the poor, young, homeless, hungry, and vagrant.  As Woody Guthrie, the Okie bard, sang of California, “Believe it or not, you won’t find it so hot, if you ain’t got that do-re-mi.”  That was in 1935, the year Cannan died; it remains true, only moreso.

Cannan goes on to say that if we are to tax site values, the tax should be national.  It is not clear how sincere he is – his style is carping, condescending, elitist and unsympathetic.  Still, his logic implies it, and he does say it, however grudgingly[4].  On this point the great Alfred Marshall agreed, in a positive spirit (positive, that is, for Marshall, a famously “two-handed” economist).[5]

Why Heed Cannan?

            It would be easy to dismiss Cannan, a careless writer.  One could pick at his many flaws, but it would be tedious and petty.  He lacked much standing in the profession, except as a hanger-on.  He is best known for editing The Wealth of Nations, the work of another man’s genius.  Marshall credits Cannan as one of many who have helped him on “special points,”[6] yet Cannan misquotes and misrenders Marshall so badly one doubts if he ever finished reading Marshall’s Principles, with its emphasis on the distinctive qualities of land, and its virtues as a tax base.[7]

Yet it would be wrong to dismiss Cannan without heeding the crash of his siege-gun, for he aimed it well.  His point is that if we are to think globally we must also act globally, or at least nationally, not just locally.  Those who follow the behest to “Think globally, act locally” trap themselves in an anomaly, dooming them to the fate of Sisyphus.  No locality has much incentive to share its land, unilaterally, with the rest of the world’s mobile people.

Alfred Marshall seconds Cannan’s point, although he notes that the “well-to-do” tend to move to the suburbs, leaving the “working classes” in central cities.[8]  He rather misses Cannan’s point that the “London Dukes” who owned (and still own) the best of central London are the target of land taxers.  At this point Marshall minimizes the problem – his world tends to be the best of all possible ones.

The Balkanized Tax Base

Differences among city tax bases are actually, however, extreme.  Parlier, a desperate little farm town in Fresno County, has just $10,000 of assessed value per head.  Here are some assessed values per head from different California cities in The County of Los Angeles: Lynwood, $21,500; Beverly Hills, $294,000 (13 times Lynwood); City of Industry, $5,533,000 (257 times Lynwood, and 553 times Parlier)[9].  Destitute Slab City (Unincorp.) in Riverside County has no land values at all.  (It is an abandoned military base between a bombing range and the fragrant southern end of the eutrophied Salton Sea, with rotting algae and dying fish.)   One would not expect much support in The City of Industry for a proposal to share land as common property with the transients who park in Slab City, which has no public services except a species of public schooling (paid by the County), nor would we expect the transients to stay in or return to Slab City if they could park on the streets of Beverly Hills, camp in its parks, attend its schools, and beg or “work for food” on Rodeo Drive.

This is why some critics have called the property tax “regressive.”  Balkanization of the property tax gives some plausibility to the otherwise bizarre claim that switching to a sales tax is less regressive than sticking with a property tax.  Within each city the property tax is progressive, but when your data meld cities like poor little Parlier and Lynwood with Beverly Hills you sometimes find poor people paying more of their income in property taxes than rich people, and getting less for it.

Then there are resource tax enclaves.  Hydrocarbons and hardrock minerals are unevenly distributed, geographically.  McLure tells us that the Siberian oblast of Tyumen, with 2% of Russia’s people, yields 65% of Russia’s oil.[10]   There are similar regional disparities worldwide.

Rich farm counties are not, generally, resource tax-enclaves (except by comparison with poor farm counties).  The “rural” counties today with high values per head are resort counties, like Vilas and Walworth in Wisconsin, with their prized lake frontages; or “exurban” counties like Napa in California; or Berkshire in Massachusetts.  In California, you might think that fruitful farming counties like Tulare have a lot more taxable real estate value per head than urban ones.  Such is a durable belief, but it is wrong.  Tulare County reports assessed values per head of $38,100.  The whole state averages $60,000 per head.  Suburban Marin County weighs in with $95,400; urban Los Angeles County has $59,000; Orange County has $74,000.[11]

You might also think that Tulare, being rural, has a higher fraction of land value in its mix, but again, not so, going by State-equalized assessed valuations.  The Land Share of Real Estate Value (LSREV) in Tulare County is 28%, compared to a statewide mean of 40%, and 47% in Orange County.  Grazing and mining counties like Inyo have high values of LSREV, but they are a small share of the farm economy.  Counties with intensive working farms, like those of the San Joaquin Valley, have low values of LSREV.[12]

Switching just the local property tax to land ex buildings will do little to correct such disparities.  It will therefore make little progress toward overall distributive justice, and the wide support that would evoke.  There is, in fact, a natural cap on local property tax rates imposed by local particularism.  The City Council of Beverly Hills will not raise land taxes in Beverly Hills to help voters in Parlier and Lynwood move to Beverly Hills and share the rents.

Local Particularism Caps the Property Tax Rate

Everything above points to there being a low ceiling on Georgist taxation applied locally.  Henry George recognized that the power elite of landowner/employers use Malthus’ doctrine to oppose raising wages – it would just spawn an invasion of new brats into the work force, they said, bringing wage rates back down to bare subsistence.  To make his points, George had to refute Malthus.  George’s view mostly prevailed, with exceptions, until fairly recent times.  Neo-classical economists even hijacked it, with a reverse spin, to trivialize land values.  Whatever we may think of Malthus today, there is no doubt that the fear of population increments from outside the taxing polity now plays the role that George ascribed to Malthusianism, and plays it with devastating effect.

Meantime, while academicians bandied words, many applied politicians saw Cannan’s Law clearly, and used it to further their ends.   The authors of the U.S. Constitution, all landowners and mostly large ones, arranged for that document to block direct Federal property and land taxes, unless the taxes be proportioned to state populations – a crippling provision.  They allowed property taxes at state and local levels – even encouraged them by blocking interstate tariffs, then the most common alternative form of revenue.  They also guaranteed free interstate migration.  Thus they assured that local particularism would cap land tax rates, while local fiscal preemption would obstruct Federal use of property taxes.  The Federalist Papers suggest that was a conscious objective.[13]  Possibly Madison and Hamilton were forced into this position to win the support of the majority of landowner-delegates, but it was they who left their fingerprints on The Federalist Papers.[14]

Austen Chamberlain, an English politician who (with his half-brother Neville) battled against proposed national land taxation from 1920-38, formulated the Tory strategy thus:

It is certain that if we do nothing the Radical Party will sooner or later establish their national tax, and once established in that form any Radical Chancellor … will find it an easy task to give a turn of the screw. … On the other hand if this source of revenue … is once given to municipalities, the Treasury will never be able to put its finger in the pie again, … [15]

Parliament followed his lead, and thus set the stage for repealing Snowden’s national land tax (it was enacted in 1931, but died aborning).  Poor Neville Chamberlain was to be the goat of such penury when he had to let Hitler humiliate him, but meantime English landlords were spared paying taxes for any national purpose.

Upton Sinclair’s 1934 run for Governor of California on the radical EPIC platform, with strong Georgist elements, was winning until the enemy found the formula of anti-Okie-ism.[16]  Jackson Ralston, running single tax initiatives in the same decade, lost to the same force redoubled, for he based his campaign on “Home Rule” for cities.[17]

Evanescent Local Successes and their Failings

There have been many temporary and partial political successes, applying Georgist ideas locally, in spite of Cannan’s Law.  These are something like correcting bad vision using eye exercises instead of glasses.  There are enough minor successes, after heroic efforts, to lead us on, but only to frustration.  Local action alone cannot achieve the main goal.  Here are a few such stories.

Some successes entail barriers to immigration.  Alaska early on set out to limit its social dividend to citizens with five years prior residence in Alaska.  It immediately lost out to the ghost of Madison.  In Zobel v. Williams (1982)[18] the U.S. Supreme Court  called this provision a barrier to interstate migration, and struck it down.  Alaska’s annual oil dividend survived, but were it not for Zobel might be much higher than today.  Meantime, Alaskan landowners pay no property taxes.  There goes much of  the dividend, and Anchorage is the most sprawled city in North America.

Significantly, exclusionary zoning has NOT been ruled a barrier to interstate migration.  Neither have state and city commuter taxes that tax the income of people who live in one state and work in another.  It may depend on whose ox is being gored.

Ethnic political machines tap into local rents while restricting the benefits to a closed circle that is hard to enter.  Their role in urban American history is well known.  So are their shortcomings, which need no belaboring here.  Note, though, that many machine politicians – Al Smith is the poster boy – have been friendlier to Georgist reforms than have patrician “good government” reformers.

Theocracies with a religious test for entry are noteworthy.  Two obvious cases are Congregationalist New England of the 17^th Century, and Mormon Utah of the 19^th Century.  Each was marked by egalitarian sharing of rents among the faithful.  Neither was able or wanted to expand its example to encompass other faiths, however, except via conversion.

California has quite a history of taxing land for public benefits.  But what public?  California cannot exclude U.S. citizens directly, but does so indirectly by winking at the widespread use of illegal alien labor for stoop and sweatshop work.  These aliens repel eastern U.S. immigrants, while the aliens, mostly non-voting, are excluded from most public benefits.

Another set of successes came from selling voters on the gains from growth and immigration.  Henry George was apparently elected Mayor of New York City in 1886 (although counted out).  He had Irish support, but was not selling an ethnic machine -Tammany and the Irish Catholic hierarchy turned against him.  He preached on the benefits of growth.  Immigrants would not dilute rents as much as they augmented them, said George.  It is a central point he underscores in his major work, Progress and Poverty.

Edward Polak (1915), a George supporter in The Bronx Borough, repeated George’s argument in supporting the proposed exemption of buildings in New York City – an exemption that was implemented, 1922-32, with a strong boost from Governor Al Smith.[19]  Now, however, there is a visible loss of belief in economies of scale of population – except in dying towns whose people feel their loss keenly, too late.

George also brought out a countervailing point that Cannan, in his exclusive concern with protecting high central rents from invasion, overlooked.  Taxes on the use and improvement of marginal lands sterilize them, said George, “and tend to drive population and wealth from them to the great cities”  (George, 1890, “The Single Tax: What it is and Why we Urge it,” in The Christian Advocate).  Godfrey Dunkley argues convincingly that that is what VAT did, when South Africa adopted it for the very purpose of making marginalized blacks pay taxes.[20]  That is not the last word on the subject either, but shows there is more to it than Cannan began to disclose.  As George maintained, aborting rent on marginal land, not just rent-sharing on superior land, distorts locational decisions.

Chambers of Commerce and Real Estate Boards have generally followed the same tack as George, touting the gains of growth.  In the single-tax era in western Canada, that crested ca. 1919, organized real estate people were a major force promoting the exemption of buildings.[21]  They often support land tax increases: some of them even opposed Proposition 13 in California.   They recognize the role of infrastructure in promoting economic development, and the benefits of untaxing buildings.  Chambers of Commerce, however, now put much more emphasis on attracting capital than labor.  Changes in fiscal federalism, discussed below, have reshaped their incentives and attitudes.

Public universities have been a screening device attracting an especially desired form of immigrant.  Local support for education is, however, lopsided, overbalanced for graduate and technical education.

In sum, local growth-orientation has become too weak, partial, and spasmodic to overcome the restrictive force of local particularism, which today dominates policy almost everywhere.  The resulting exclusionary policies, when practiced by all or most localities, drive landless proles from pillar to post until they become so desperate they will serve landowner-employers for very little.  It is not enough to “think globally”: we must act globally.  “Some for the Glories of This World, and some/ sigh for the Prophet’s Paradise to come; …”.  Now, it seems, to win some glories of this world we must do more than just sigh for the Prophet’s Paradise, we must work for it.

Acting Globally

One way to act globally (or at least nationally) is through a national land tax, or some reasonable facsimile thereof, coupled with a national citizens’ dividend.  The income tax act of 1894 did include land income in the tax base, thanks to the persistence of a handful of single-tax Congressmen – yes, really, there once were such men, six of them at that time.  The U.S. Supreme Court struck it down because property income was in the base, [22] but President Taft (of all people), Congress, and the voters came back with the 16^th Amendment, adopted in 1913, that did include land income in the tax base.  When Congress, led by single-taxers Warren Worth Bailey (of Johnstown, Pa.) and Henry George, Jr. (of Brooklyn), first implemented the amendment it virtually exempted wages and salaries by exempting incomes below a high cutoff point.[23]  The brunt of federal taxation fell on property income, much of it land income, and it was enough to finance World War I.

Since then the income tax has evolved, step by step, into its present anti-labor form, with most property income exempt de facto, and high rates on earned income.[24]  It is obviously constitutional to reverse that trend, because we have been there before.  It would also be desirable, but here we will focus on the cognate matter of “fiscal federalism.”

To enable basic tax reform at the local level we must deal with local particularism.  To do that, in turn, we must deal with “fiscal federalism.”  How are central governments to distribute funds from their so-called “surplus”: to people (as a social dividend), or to local governments representing landowners?  When we wake up to smell this coffee, we will find that a lot of economists have gotten up first.  Many of these economists deal with LAND RENT, defined as Ricardo would.[25]

The reason it is so hard to sell growth policies – like land-value taxation – at the local level today is that fiscal federalism, as practiced today, is perverse.  Central governments, imbued with the anti-personnel spirit of Austen Chamberlain, tax people as people, while handing out subventions to landowners as such, and to local governments as such.  The landowners can get the subventions without having people, so who needs people?  That’s our problem in a nutshell.  Persons as such become fiscal pollutants, from the local view.  After the T-Men have plucked their feathers, working persons are less able to pay local taxes; while Federal grants relieve local landowners from needing population to share public costs.[26]

Perverse fiscal federalism is DEsocialization of rent – creating new private rents using public monies wrung from workers.  This is inherent in grants for capital spending,  e.g. for sewerage; and tax exemption of muni bonds.  These grants and exemptions are given to municipalities as such.  That is only a step away from returning dollars to landowners as such, because municipalities are defined as areas of land, a group of local landowners.[27]  Desocialization is inherent in farm subsidies, e.g. payments to fallow land, using tax money from workers.  It is inherent in preferential assessment of farmland, e.g. California’s Williamson Act, where the state pays localities for their lost tax revenues from underutilizing lands.  It is inherent in the use of property-tax exemptions to subsidize many underutilizations of land and hobbies of the rich, like redundant airports for private jets, cemeteries, golf courses, campuses, church parking lots, conservation easements, timber, etc.  Some of these may foster socially defensible uses, but note it is the lands, not the personnel, that are tax-exempted.

Canada’s classic Carter Commission Report[28] led the right way, but Canada’s actual equalization program leads the wrong way.[29]  Equalization grants from Ottawa to the provinces are lower to provinces whose taxable capacity per head is higher, and of course vice versa, according to a detailed formula.  So far, so good, but the devil is in the definition of “taxable capacity.”   Canada specifically excludes land value from measures of taxable capacity.[30]  Buildings are included as part of the potential tax base; a hardworking productive population is included; a thriving commerce is included; but land value is quietly excluded.  Thus a province wherein vast and valuable lands are underused is considered a charity case, eligible for alms from Ottawa; while another province that makes productive use of meager lands has to pay more taxes, but gets less relief.  That helps explain why Ontario and Quebec, despite their great urban and locational advantages, still rank below the provincial average in measured taxable capacity.[31]  It is not the capacity that is lacking, but the measurement of it.  The tilt is patent; it could hardly be an accident.  If any one of the many brilliant economists, politicians, and bureaucrats who prate or publish on equalization payments, horizontal fiscal federalism, and Canada’s Representative Tax System (RTS) has even peeped on this point, I am not aware of it.  Their consciousness has fallen below the threshold of perception, and needs desperately to rise.

Within provinces there are equalization programs, too.  British Columbia offsets the magnetism of Vancouver by subsidizing less magnetic cities from general revenues, and by cross-subsidizing rail and utility services to distant outposts in the boonies, but it is local governments or private landowners, not people as such, that get the benefits.  It is the same in every American state.  The exception is public education, which is therefore the target of the most spirited attacks by privatizers (like smug George Will) who dominate the op-ed pages today.

The modern “Public Choice” school has grown terribly chic in the economics profession.  It focuses on fear of  “the tyranny of the majority,” given votes.  The basic concept is unrealistic and prejudicial, in view the observable fact that the minority of landowners, armed with discretionary wealth, sway the majority of voters to support policies that favor landowners over the underlying population.   The Public Choice school leads us to fear and fend off an imaginary problem, blinding us to the real one that is quite the reverse.  Veblen explained voter behavior better by analyzing the mindset of voters as a cultural throwback to an age of marauding Viking bands organized around mindless fealty to some alpha male, whom the betas and omegas were bound loyally to support and serve at any cost to themselves.  A progressive society must learn to place more value on the “instinct of workmanship,” and express its unity in more egalitarian ways.

Reversing Perverse Policies

Public spending should feature “Citizen Dividends.”  These are social dividends limited to citizens, thus discouraging free or illegal immigration that would dilute the dividends and erode their voter support.  (The degree, pace, and conditions of legal immigration is an issue to treat separately.)  Dividends take many forms other than outright per head cash grants.  The G.I. Bill was a splendid example.  Social Security payments are another.  School equalization payments based on average daily attendance (a.d.a.) are another.  A state or province cannot easily restrict benefits to its old time citizens, as Zobel showed – but a nation can.

At the same time, there should be no more capital grants to localities for public works.  When cities pay for their own public works they must attract population to justify the capital outlays and service the debt.

Federal taxation should bear heavier on land income, and lighter on wage and salary income, as in 1916.  It was constitutional then; it still is.  The combination of a citizens’ dividend and income-tax reform would drastically rebalance local incentives.  Cities would compete to attract median people rather than, as now, to repel them.  This would not cause swamping of cities with people because it is a zero-sum game in a closed system.  Competition would simply raise wage rates and lower living costs.[32]

Congress should repeal the tax exemption of state and local bonds, a massive ongoing subsidy to local landowners.  This repeal will be challenged as an invasion of state sovereignty, but recall that Congress had no trouble in 1939 repealing the tax exemption of state and local employees.  Would the courts find bonds to be more sacred than payrolls?  To find out, we only need a simple act of Congress that would quickly be adjudicated.

The federal government should review local zoning, and other exclusionary policies, as barriers to interstate migration.

There is a federal interest in better tax assessment of land, to keep buyers of used buildings from overallocating their tax “basis” to depreciable buildings, thus arranging falsely to depreciate land, and erode federal revenues.  Something like a national board of equalization is called for.  The U.S. Census of Governments, with the pioneering work of Allen Manvel and political support from Illinois Senator and Economics Professor Paul Douglas, established the precedent.  While we’re at it, let us outlaw the sequential depreciation of the same building by successive owners, an obvious outrage.

The result of such measures would be to restore the concepts of dignity of labor, and the key role of income-creating investing (as opposed to acquiring existing wealth and rent-seeking).

Colin Clark’s national land tax

For nations where a national land tax is politically thinkable, Colin Clark[33] has proposed a simple technique to spike Cannan’s big guns.  Says Clark, “… land values per head of population should first be ascertained; then the state would impose a land tax which exempted altogether those local authority areas where per-head land values were low, and which rose in a progressive scale for those with higher land values per head.  Each local authority would then also impose its own tax, …”

Alfred Marshall, disguising his boldness under a mousy writing style, proposed an even stronger supplement to the land tax.  He would make the tax base the capital value of land, rather than the annual cash value, to tap “the part of the real annual value of land which does not appear in a money form … ”.  Repeating himself for emphasis, he says that taxing capital value will “bring under taxation some real income, which has escaped taxation merely because it does not appear above the surface in a money form.”  That is, Marshall wants the national tax to fall on imputed land income, an enormous annual flow of value that now totally escapes income taxation.

And what is the value of land under old buildings?  Marshall writes no nonsense about seeking the depreciated value of the old building first.  Land value is the opportunity cost of the site itself: what land would bring “if cleared of buildings and sold in a free market.”[34]    Imagine how that set of policies, from this prissy pillar of property and propriety, would radicalize national taxation in any modern state.  Beneath the cautious façade, Marshall reinforced some Georgist ideas.

Yet there is more.  Marshall applauds Lloyd-George’s “Social Welfare” Budget of 1909, the one that humbled the House of Lords, because the proposed land tax will “check the appropriation of what is really public property by private persons.”[35]   Did Henry George ever say it plainer, or more provocatively “in-your-face”?  No wonder Edwin Cannan shied away from mastering Marshall’s Principles.  No wonder George Stigler had to go back to a disorderly altercation at Oxford, and alleged comments that Marshall never published, shiftily to define the great Marshall as an anti-Georgist.[36]  Can we, in our Federal system, come up with something comparable to the ideas of Clark and Marshall?  It is a matter of thinking creatively, with the right attitude.

Our worst enemies can be our best friends, when we learn from their criticisms.  Cannan’s shot found and breached a weak spot in the Georgist line, and bade Georgists fall back and regroup.  So long as modern Georgists ignore and dismiss the Cannanade, they will continue to suffer for it.  When they analyze their setback and learn its lessons, they will advance to their goals.