I have a new paper out, looking at the impact of stamp duty on the housing market. Methodologically, the question turns out to be slightly tricky – because stamp duty is a mechanical function of house prices, a regression of turnover or prices on average stamp duty in a neighbourhood would return a positive coefficient. To get the behavioural impact, it’s necessary to form a measure of the predicted stamp duty in an area, and run this either as a reduced-form or instrumental variables regression. When you do this, the standard predictions of economic theory are borne out: places with higher predicted stamp duty rates tend to have lower housing turnover and lower house prices. Abstract below (click on the title for the full paper).
How Do Stamp Duties Affect the Housing Market?
Land transfer taxes are a substantial portion of the cost of moving house in many developed countries. However, little is known about the effect of such taxes on the housing market. Since stamp duties are endogenous with respect to the house price, I create an instrumental variable that is the stamp duty on a property, given that postcode’s starting house price and the national house price trend. In a specification with postcode and year fixed effects, this instrument effectively captures policy changes and nonlinearities in the stamp duty schedule. I find that the impact of an increase in the tax rate is to lower house prices, with the magnitude of the effect rising slightly over the medium run. I also observe impacts of stamp duty on housing turnover. A 10 percent increase in stamp duty lowers turnover by 1-2 percent in the first year, and by 4-5 percent if sustained over a 3 year period.
Of course, just because stamp duties affect the housing market, it doesn’t automatically follow that we should scrap them. My quick rundown of that issue here:
Against stamp duties:
- Internationally, Australia’s current rates are pretty high (an easy way to think about it is that stamp duty on a given house is about the new price of the car in the driveway).
- Higher mobility taxes mean that you get more mismatch between families and the housing stock (too little upsizing by young families, too little downsizing by empty-nesters).
- Higher mobility taxes mean that people may not take a job in another city (leading to lower productivity, or lower employment). Or it could mean that they live in the ‘wrong’ part of town, and waste too much time and petrol commuting.
In favour of stamp duties:
- All taxes have their distortions, and we’d want to weigh up the those imposed by stamp duty against the distortions that are a byproduct of sales, corporate or personal income taxes. Land taxes are regarded by economists as pretty efficient, and some experts argue that if large land taxes are politically infeasible, land transfer taxes are a good second-best.
- Residential stability leads to higher social capital in neighbourhoods. People ignore the negative externality that moving imposes on their neighbourhood’s social fabric, and should therefore be taxed for moving.
Core Economics 
Andrew,
You state that stamp duties might be justified because they reduce the incentives to relocate which reduces social capital. This sounds like an extremely iffy conclusion since these externalities cannot be measured and seem a political value judgement rather than an identified distortion.
I also think your conditional view that the finding that stamp duties reduce house prices does not provide a case for their abolition would be rejected by almost all tax economists in Australia. These are among the most foolish state taxes in a large sparsely populated country where labour mobility is of crucial importance in maintaining a semblance of labour market efficiency.
Indeed we are forced to rely on immigration as a way of dealing with immobility because of these high taxes.
We would be much better off giving the states back a part of the income tax than relying on these foolish taxes.
And why the need for empirics on this. People bid for a house and they do not consider the stamp duties? This seems impossible to berlieve.
LikeLike
Harry, I confess that I thought my comments were utterly unexceptional. Theory can give us signs, but we typically need empirics to get magnitudes. And once we have those magnitudes, we need to compare them with other distortions (and think about externalities) before making a policy decision.
LikeLike
Andrew
I wonder if the empirics give better forecasts of the effects of these taxes than <i>a priori </i> reasoning. There are measurement issues in calculating house prices and the effects you calibrate seem low.
If I am bidding for a house for $400,000 why wouldn’t my bid reflect the effects of a sales tax of say $28,000 I know I must incur as a consequence of purchasing? Apart from some quirky behavioural economics reasoning the tax should be reflected 100% in a reduced offer.
My point about externalities is that there are strong efficiency costs in limiting the ability of workers to relocate or allowing parents to relocate to a smaller dwelling when their progeny leave home. These have been discussed for decades – they are some of the worst ways for state governments to raise revenue. To recognise these but to suggest the argument against such taxes is non-compelling on the grounds of reduced social capital consequent on greater mobility seems to muddy the waters unnecessarily.
Again general economics reasoning suggests these externality costs will be less than the direct gains people enjoy from having greater flexibility. If I have good neighbours and enjoy interactions with them and they with me, fair enough. But are you suggesting that people should be discouraged from moving to suit themselves (their surplus gains less their lost social capital) because those they leave behind will be worse off? This does not sound a robust view.
LikeLike
Andrew the other inefficency to remember is that lots of houses tend to get “over-renovated”. Given the stamp duty cost, lots of mid-size houses in expensive suburbs tend to have large reno’s done since its cheaper than buying a bigger house. Rather than buying a 4 bedroom house its cheaper to just Renovate a 2-3 bedroom.
In victoria stamp duty in $1M is $55K. That goes a fair way.
LikeLike
Empirically, the decision to move or not is based less on the level of stamp duty than on the nominal purchasing price itself. Do the results vary if stamp duty is paid after sale, rather than at purchase?
If you wish to observe social mobility in action in relation to pricing and taxes, a country where they have a good mix of home ownership and long-term rentals (with societal norms that do not favour home ownership) provides an excellent test bed for barriers to exit and entry. People utilizing long-term rental contracts are, as a rule, highly mobile (and usually, but not always younger), whereas those undertaking the option of home ownership tend to invest socially and emotionally in where they live (and typically have families).
LikeLike
I suspect a small error in your conclusion Andrew. You state that the tax incidence is on the buyer, but if price declines caused by the tax are greater that the tax, wouldn’t the tax incidence be on the seller?
LikeLike
Cameron, you’re quite right, thanks (hopefully the rest of the paragraph made the meaning clear). Sentence now reads: “these results imply that the economic incidence of the tax is entirely on the seller; that is, prices fall by the full amount of the tax.”
HC, I don’t think incidence tells us anything about efficiency. That is, a tax might be fully passed through, but cause minimal distortion. We need empirics to work out the impact of taxes on turnover.
DP, the country that’s closest to what you describe is Switzerland, from what I know. Not sure if people have looked at mobility taxes there.
LikeLike
Andrew,
In your findings you imply that “that the economic incidence of the tax is entirely on the seller; that is, prices fall by the full amount of the tax.”
May I ask then: –
1. At what level would that tax simply prohibit all movement and home ownership aspirations.
2. If you were able to use a sliding scale of duty and study the effect over multiple levels, is there a “tipping point” where the level becomes self defeating by starting to reduce the sales volumes to such a degree that state government revenue from duty falls away?
Your research seems to draw the conclusion that any level of duty will be deducted from the purchase price, and thus the buyer is not disadvantaged by the duty, but there will be a point of resistance.
3. What does your research indicate in that area?
LikeLike