An exciting email pinged into my inbox at the end of last week. It was a link to the contents of the latest issue of the American Economic Association’s journal Economic Policy. For most people, these publications are not usually as gripping as, say, a Ken Follett novel. But nestling among the thickets of algebra, there was an article entitled “Mansion Tax: the Effect on the Residential Real Estate Market”.
The authors, Wojciech Kopczuk and David Munroe, are both members of the prestigious economics department at Columbia University in New York. The article contains its fair share of technical material, but the main points are conveyed by some straightforward charts. The paper describes a detailed analysis of residential real estate transactions since 2003 in New York City and in the neighbouring state of New Jersey.
The taxes on high value property which they examine are nowhere near as punitive as the annual levy envisaged by Ed Miliband. But their impact has been both dramatic and detrimental. A so-called mansion tax has been in force in New York since 1989 and in New Jersey since 2004. It applies not on an annual basis to the property, but simply to transactions of $1m and over. The tax rate is 1 per cent and is imposed on the full value of the property – so that a $1m sale is subject to a $10,000 tax liability, while a $999,999 transaction is not subject to the tax at all.
The tax distorts the distribution of pricing so that there is a large bunching effect just below the $1m threshold. Unsurprisingly, a plot of the number of transactions in New York City against the price slopes downwards. There are a lot more sales of properties at, say, $500,000 than there are at $1.5m. But there is a huge spike in the chart immediately below the $1m tax threshold. The same result is shown for New Jersey. Here, the tax was introduced during the period for which the transactions data is available. And its impact was virtually instantaneous.
There is a large gap in transactions in price bands just above the threshold. Crucially, this is bigger than the excessive number of sales which take place below it. More sales are lost above the threshold than are gained below it. In short, the tax causes the market to, as the authors put it, unravel. It ceases to function properly, and trades which would otherwise have been undertaken by willing buyers and sellers do not take place at all.
This is the real subtlety to the article. Mansion taxes create costs which are not at all obvious at first sight. The people who lose out can never be identified, for the simple reason that they are unable to carry out the transactions which they would have liked to. But their losses are nevertheless real.
Markets are often imperfect instruments. If left completely unchecked, they can create undesirable outcomes. But Soviet-style diktats on markets can create costs for society which go beyond the immediately obvious.