Double stamp duty
25 February 2013
The level of economic illiteracy behind Friday's proposed doubling of stamp duty rates is breathtaking. These are proposed measures, which will proceed only if the Legislative Council allows them to. So although we covered this briefly in our alternative budget speech on Friday, it is still worth explaining why these measures are misconceived, in the hope of educating decision-makers.
Not demand management, but volume management
First of all, there is a cognitive bias caused by the fact that although both parties are jointly liable for the stamp duty on a transaction, the duty is normally paid by the purchaser. This results in the misperception that increasing duty rates deters purchasers more than sellers, because "the buyer pays". The Government even calls it a "demand-side management measure", strengthening the misperception. You can overcome this misperception by realising that stamp duty is equivalent to a brokerage charge by the Government on both sides, charged on the mid-point between what the buyer pays (including stamp) and what the seller receives (excluding stamp).
In reality, the deterrent effect of increasing stamp duties is on both sides, locking sellers in and locking buyers out. It is not demand management, but volume management, particularly in non-residential property where there are no exemptions.
Bear with us for a bit of algebra. If "H" is the headline price, upon which stamp is paid, then the seller receives H, while the buyer pays H+D, where "D" is the duty in dollars. Half-way between these is the "mid-price" which they would agree upon if no duty was payable, being P=H+D/2. For example, if a transaction at H=$6m was stamped at a rate of R=3%, then D=R*H=$180,000. The buyer pays $6,180,000, and the seller receives $6,000,000. The mid-point, P=$6,090,000 is the "real" price, unaffected by stamp duty rates. The stamp duty is equivalent to a Government brokerage charge on each side of 1.47783%, or $90,000. To save you the algebra, the general formula for that brokerage rate is B=R/(2+R). In this example, B=0.03/(2+0.03)=1.47783%.
On Friday, the top rate, on properties over $20m, was doubled to 8.5%. That is equivalent to a brokerage of 0.085/2.085=4.08% on each side of the mid-price.
Increasing stamp rates are neither a "demand-side measure" nor a "supply-side measure". They are simply a "volume suppression" measure. As the cost of transactions increases, the number of transactions declines. If, for example, stamp duty were increased to 100%, then the effective brokerage would be B=1/3, or about 33.33% on each side of the mid-price. Very few transactions would get done, because the government would in effect be confiscating 2/3 of the value of anything which changes hands. And if duty was increased to 200%, then the confiscation would be exactly equal to the mid-price. In our example, at 200%, the seller would receive $3,045,000, the buyer would pay $9,135,000 and the Government would collect $6,090,000 in stamp. Here's an expanded table which shows the effect, if any transactions were done at higher rates:
So higher rates suppress volume, but they have no effect on mid-prices. You get a reduction in headline transaction prices (excluding stamp), but when you add back half of the stamp duty, you are back to the mid-price. The reduction factor in the headline price is simply (2+R1)/(2+R2) where R1 is the old rate and R2 is the new rate. So a doubling of stamp on office properties from 4.25% to 8.5% will reduce headline prices by a factor of 2.0425/2.085, or about 2.04%. Each side will be worse off by the same dollar amount, being half of the increase in duty paid. So when you see headlines which say "office prices drop 2% in wake of new measures" you will know that the mid-price is unchanged and the illusion is complete.
Friday's announcement contains an exemption from the increase for first-time home buyers who are permanent residents, but there is no exemption for buyers of offices, shops, car parks, hotels, industrial units or anything non-residential (except for redevelopers). The higher costs will deter potential sellers and buyers, but do nothing to affect HK's stock of floor space or its economic value. Instead, many potential sellers will lease out their units, and many potential buyers will rent instead. The potential seller has his capital trapped, and the potential buyer has to tolerate the regular renegotiation and relocation that comes with being a tenant. Both are deterred from exercising their rights in Basic Law article 105 to acquire and dispose of property, including real estate.
All that the Government will be able to show from this is a reduction in transaction volumes, and an illusory reduction in transacted prices by an amount equal to half of the increase in stamp duty. The reduced flexibility for potential sellers and buyers just throws extra sand into the wheels of commerce. If that is the Government's objective, then they have succeeded.
There is a concept in income taxation called the Laffer curve which plots revenue against tax rates, and it is similarly applicable to transaction taxes such as stamp duty. As the rate increases, the volume decreases. Starting from zero rates, revenue rises to a peak and then declines back to zero. We suspect that after Friday's increase, we are already past that maximum-revenue point, particularly for higher-value transactions on properties without exemptions. In other words, when you double the rate on offices over $20m from 4.25% to 8.5%, the combined transaction value will drop more than 50% and stamp revenues will decline.
With high enough rates of stamp duty across the board, we could bring transactions to a halt. That would not mean that properties are worthless, only that they can no longer be transferred. Those who need space will rent from those who own space - rents won't be materially affected.
The other effect would be more work for lawyers to document transfers of companies which own property in HK. Many properties, particularly whole buildings and larger premises, are already held by single-asset companies, so they will effectively be exempt from stamp duty anyway. If shares are registered in HK, they are subject to 0.2% stamp duty on net asset value (net of any mortgage or loans), but if a HK company is owned by an offshore company, then the offshore company can be transferred without paying any stamp duty in HK. So the biggest impact of the new measures will be on small and medium enterprises who just want to buy or sell an office unit, shop or industrial unit for self-occupancy.
How widespread is corporate ownership? The Chief Executive Leung Chun Ying himself owns his Peak houses via Lotvest Limited (HK). He could sell that company whenever he wants. Secretary for Development Paul Chan Mo-po's family owns a flat in HK via Victorwood Development Limited (HK), which they could also sell. Your editor owns his flat via an offshore company which paid stamp duty on the property purchase. The register of interests also shows that several unofficial Executive Council members own properties via offshore companies.
What does "cooling down" mean?
In Friday's press release, the government conflates transaction volumes with prices, leaving us unsure what they are trying to achieve. If by "cooling down" they mean reducing transaction volumes, then increases in stamp duty will certainly have that effect, although it is hard to see how that benefits anyone. But if by "cooling down" they mean reducing mid-prices, or reducing the velocity of mid-prices (the rate of change over time), then they will have no effect. Capital values of properties are driven by discount factors and expected future net rental income. Rents reflect the balance between supply and demand for premises, and discount factors reflect expectations of interest rates and appetite for risk. None of these is affected by the rate of stamp duty per se.
If there is any effect at all, it is only that the measures increase the perception that the Government no longer respects the rule of law and Hong Kong's free market system, which was supposed to be unchanged for 50 years after 1997. The expectation of future meddling may itself increase discount rates and reduce the attraction of investment in Hong Kong's economy. That uncertainty would reduce values, but trashing HK's reputation as a free and open economy is an expensive price to pay for manipulating the market.
So let's see if we can reduce this to 5 points that every minister or legislator can understand:
- Stamp duty is equivalent to a brokerage charged by Government to both sides, at the mid-point between the price excluding and including stamp duty
- Higher stamp duty rates are equivalent to higher brokerage charges on both sides. They are not a "demand-side measure"
- Higher transaction costs reduce transaction volumes and economic flexibility, but don't affect economic values
- Beyond a certain level which we have probably just passed, higher stamp duty rates reduce Government revenues
- Meddling with the economy will damage confidence in Hong Kong.
© Webb-site.com, 2013