Comments on a Study of the Taxation
of Land Values in Pennsylvania

Mason Gaffney and Dan Sullivan

[Reprinted from the Land-Theory online discussion group, 1999]

Mase Gaffney:

I note the following:

The authors are thoroughly right to stress the importance of assessment practices. These may be as important as, or even more important than formal legislation. However, the pages spent on the Bentick-Mills quibble might then better be spent on the more important general question, how does the assessor divide land from building value? There are two basic concepts: the building-residual method, and the land-residual method. The authors are silent on this matter. LVT clearly presupposes use of the building-residual method. You value the land as though clear, and allot whatever remains to the existing building. Cf. #24, below.

Dan Sullivan:

While assessment practices are important, and while they are bad in Pennsylvania, there was no significant change in assessment practices that coincide with Pittsburgh's construction increases so as to form an explanation for those increases. If the purpose of the paper is to examine possible causes of construction other than the land value tax, discussion of assessment practices is wholly digressive.

Mase Gaffney:

The authors are wrong to attribute all the effects of LVT to the abatement of other taxes. They dismiss liquidity effects by citing a 1990 study of Steven Bourassa, p.8. That is hardly enough to overcome the presumption created by the galloping spread of preferential assessment of farmland around growing cities; of timberland; and of golf courses (in California). The entire rationale behind preferential assessment, from its inception in Maryland in 1957 under Gov. McKeldin (spearheaded by Assemblyman Spiro Agnew), is the universal perception that higher land assessments induce conversion of land to higher uses (coupled with a value judgement that this is bad).

Dan Sullivan:

Also, no such preferential assessments existed Pittsburgh, and they are so rare in the municipalities contiguous with Pittsburgh as to be inconsequential. There are a variety of abatements on new construction, but these exist elsewhere, without comparable results.

Mase Gaffney:

P.13, shouldn't the reader be advised that the Penna. Economy League is a political organization with constituents and an agenda; and Mayor Caliguiri led the political opposition to the graded tax plan? The authors seem to treat it as though it were a disinterested research organization, and the political Mayor as though he were of an uninvolved, judicial mindset.

Dan Sullivan:

On paper, the PEL is a disinterested research organization, and they have in fact built up considerable stature in that regard. However, the PEL study of land value tax was an orchestrated hatchet job. One of the textbooks for my college statistics course was *How to Lie with Statistics.* It showed 23 ways to manipulate the truth through statistical misrepresentation. The PEL study used 21 of them. The other two involved using graphs, and the PEL contained only one graph, which was drawn honestly. If not for that graph, the PEL study might have acheived pure intellectual dishonesty.

Some explanation is needed of why the comparison cities were chosen. Is the word "sample" chosen advisedly? Proximity and size seem to be the criteria used. Only two other Penna. [control] cities are included, Erie and Allentown, and they are much smaller. I should think Philadelphia would make an obvious pairing with Pittsburgh, because both are subject to the same State laws and tax regimen.

Philadelphia is a much larger city, and is covered under a different municipal classification, with a different tax code. Most notably, Philadephia has a commuter wage tax, which is as economically destructive as it is politically addictive. The wage tax in Philadelphia is now almost 5% on residents and almost 4% on commuters.

Cleveland and Buffalo are actually more comparable socially and economically, as they are about the same size, and are both rust belt cities, plagued by industrial shutdowns. Cleveland, like Pittsburgh, has an unusually high number of major corporate headquarters for a city of its size.

Mase Gaffney:

The impetus was much stronger for nonresidential. This is not exactly what one would expect a priori; some explanation would be welcome. Two explanations seem worth exploring. 1) Land in the City of Pittsburgh is mostly well suited to non-residential use, so there was some tendency to convert land from residential to non-residential use. 2) Perhaps apartments and hotel-apartments were classed as commercial uses? This is often done, and can confuse one's results if one fails to pick it up.

Dan Sullivan:

The explanation around here is that Pittsburgh has long had a declining population. There is no need to build residential housing when mere maintenance of existing housing would suffice.

Also, the increases in land value tax largely coincided with increases in wage taxes from 1% in 1979 to 4% by 1985. In 1988, the Masloff administration commissioned the University of Pittsburgh to find out why people were leaving the city at the alarming rate of 4% per year. Pitt researchers interviewed people who had left and found that the number one reason given by these expatriots was the city's higher wage tax. (The prevailing suburban rate is still 1%.) That year, the city cut the wage tax by 1/2%, and the following year cut another 5/8%, making up the difference mostly with land value tax. The population quickly stabilized.

Mase Gaffney:

P.20, 3rd-last line, suggests perhaps the housing performance would look better if compared with the national trend. Why not do so, rather than leave this hanging? It need not be terribly formal - just a few apposite data would do.

Dan Sullivan:

I suspect that the national trend is a tougher comparison than metropolitan, state or regional trends. Things have been slow in Pennsylvania generally since the mill shutdowns of the late 70's, while the Southwest boomed. I think the best comparisons are between Pennsylvania cities that switch to land value tax and Pennsylvania cities that do not. If we cannot compare Pittsburgh to Philadelphia, we can certainly compare Scranton to Wilkes-Barre, Duquesne to Clairton, Aliquippa to Ambridge, etc.

Mase Gaffney

P.22, the writers may be reaching too far in their interpretation of the Penna. Economy League findings. The League questionnaire apparently asked people if "the shift to heavier land taxation" (and, I presume, lower taxes on new buildings) had an effect. The writers reinterpret that to mean that the respondents meant that LVT by itself had no effect, in a sense that would be used mainly by fiscal theorists, not the respondents in question. It is cute to turn the League against itself, but I am not persuaded. However, I have not seen the survey.

Dan Sullivan

Nor will you see it, as it was oral. I followed up on this by asking the PEL interviewers and some of the interviewees what was actually said. Even the list of the interviewees was not given, and I had to persist in asking pointed questions to find out who was interviewed. Many of the interviewees were real estate executives who had already testified against the land value tax and who had a vested interest in opposing it. Even those who might have been objective were not given objective

Questions. The central question was commonly put in the form of this false dichotomy: "Do you think land value tax caused the construction boom or were there other causes?" Well of course there are other causes. One of the interviewees, professor Donald Stone, head of the School of Urban and Public Affairs at Carnegie-Mellon University, saw through the dichotomy and proceeded to talk about the beneficial effects of land value tax when the interviewer changed the subject to assessment questions. Stone commented afterward that he found it hard to believe that the interview of him was related to the resultant study.

Mase Gaffney

P.23, bottom, "some of the more extravagant claims that land-tax proponents have made ... " Cite some, or drop this. Every movement produces its windmill orators. Extravagant claims are not peculiar to land-tax proponents, and it is unfair to imply so by innuendo. Remember Arthur Laffer, Jr.? Howard Jarvis? Hall and Rabushka? Listen today to Richard Armey, Charls Walker, Rush Limbaugh, Paul Craig Roberts, Margo Thorning, or Newt Gingrich. Indeed, wild, incredible claims about the benefits to flow from untaxing capital gains have nearly gone mainstream.

Dan Sullivan

I note only conservative windmills blowing here. What ever happened to Lester Thurow, Ted Kennedy etc.? (Just tugging at Mase's liberal chain.)

Mase Gaffney

P.24. Cf. #3, above. They conclude LVT is "neutral," and therefore good. As a cautious rhetorical device, given the values dominating the profession in 1995, this may be the strongest conclusion they are wise to make.

Dan Sullivan:

Except that it does not correlate with the facts. Land value tax rates were increased in response to lost revenue from mill shutdowns, outmigration, etc. It was not accompanied by reductions in other taxes nor by increased public services. Were it neutral, it would have been expected to produce no effect whatever. Some of the the other Pennsylvania cities did reduce building taxes as they raised land taxes, but others merely responded to a disappearing tax base the way Pittsburgh did. Both categories of cities enjoyed construction increases, according to data gathered by Cord and Vincent of the Center for the Study of Economics, and by the Oates and Schwab data as well.

Anyone who thinks the land value tax is neutral in its effects only need watch the apoplectic land speculators fume at city council when the tax is proposed, for the direct effect is not so much to induce developers to come to Pittsburgh, but to induce the speculators to offer better deals.

One year, the head of the Building Owners and Managers Association argued that taxation is a minor factor in attracting major development. Thinking that he was making his case, he noted that (prior to the land value tax hikes) Rockwell International had wanted to locate their corporate headquarters in Pittsburgh's Golden Triangle (CBD), but could not find an available location. Well, that is precisely the problem that the land value tax addresses; builders of major buildings have had much less trouble finding cooperative landowners since the land value tax increases.