Twenty years ago, I wrote an article titled “Greater Vision Needed to Curb Speculation” (see SCMP 12 August 1991). I hoped at that time to explain to the public why it was not a good idea for the Government to intervene in the pre-sale market for housing units. What was clear to me then, as it is now, was that there was no compelling reason to believe the market was failing or that Government intervention to curb speculation was warranted.
Speculative activities improve economic efficiency. I had hoped that a better appreciation of the mechanism behind the pre-sale arrangements would lead to a more enlightened view about regulating the market for pre-sale housing units. To my disappointment, this has not happened, although Mr. Tang Kwong-Yiu, who later became Government Economist, congratulated me for a most illuminating article after it was published in 1991. I was gratified that at least one person in Government understood what was happening, but in the past 20 years this has not come to anything. The Government has repeatedly intervened in the market to curb speculation whenever there is a public outcry against property price surges.
Curbing Speculation has Temporary Effects
The public often believes mistakenly that speculation causes property prices to rise. And when Government intervenes to curb speculation this mistaken belief is unfortunately reinforced. In reality, speculation is the result of anticipated property price increases and it is such anticipation that drives up property purchases. This anticipation arises from underlying supply and demand conditions that lead prospective purchasers to believe prices will rise, and speculators to become active. The perception that speculators drive up property prices is an easy popular misunderstanding that has been around for a long time.
The evidence that speculation is not the main cause of surging prices can be seen in the history of Government attempts to curb speculation. Every time a measure is introduced, it is followed by a short correction and then prices surge again.
In 1991, in response to rising property prices, the Government introduced several measures to curb speculation in the pre-sale market. First, units reserved for internal sales were limited to 50%. Second, confirmor transactions in pre-sold units (i.e. forward delivery) were subject to stamp duty at the rate of 2.75% of the value. As a consequence, property prices made a minor correction before surging upwards again (see Chart). In 1992, the Sandwich Class Housing Scheme was announced and property prices again briefly corrected before continuing their upward surge.
The Puzzle of the Pre-Sales Market
In 1994, two additional measures were introduced to curb speculation in another bid to halt continuing property price increases. First, pre-sales of residential units under construction could be undertaken only within 9 months of their completion. Second, purchasers had to pay a 10% deposit instead of 5%, and a charge of 5% was made if the transaction was subsequently cancelled. Property prices corrected before continuing to surge upward again.
After 1997 some of these measures were rolled back as the Asian Financial Crisis took its toll on the economy and the property market. But after 2003 property prices began to surge forward again. There was another correction in prices in the wake of the Global Financial Tsunami in 2008, then another surge shortly afterwards. Speculation was no longer limited to the pre-sale market, but had spread to the secondary market for existing units.
The Government stepped in with a further raft of measures. In 2009, the loan to value ratio for property transactions above $20 million was lowered to 60% and tighter regulations on pre-sales units were introduced. In 2010, the stamp duty for property transactions above $20 million was raised to 4.25%. The Government also pledged to launch more land sites for sale. Later that year confirmor transactions for units under construction was banned and a 10% levy was charged on all cancelled transactions. In 2011, a special stamp duty was levied for property transactions that were sold within 2 years. These last measures effectively ended the forward housing market.The Government also launched the My Home Purchase Plan and later introduced the revised new style Homeownership Scheme.
Throughout this period, property prices continued to surge even though there was a slight halt whenever measures to curb speculation in the pre-sale market and the secondary market were introduced.
Back in 1991, my article explored the evidence, or lack of it, of speculation in contributing to market failure. There is some personal background to that article. About a year before I wrote it, I purchased my first ever property inHong Kongin the pre-sale market. It was an eye opening experience for me as I observed first-hand how the pre-sale housing market functioned. I began to follow the practices of developers through the newspapers to try to understand their rationale.
After about a year, I had enough of an understanding of how the pre-sale housing market was organized to construct an economic interpretation of the main features of its practice. Why do they choose to sell some of their units internally to speculators in advance of the announced date of sale? Why do they choose to make prospective buyers wait in queues? Why don’t they use auctions, which could ideally maximize their expected gains as it allows them to price discriminate?
Speculation is Economically Positive
Around the same time property prices were rising dramatically and there was intense popular demand for the Government to intervene to curb speculation. I believed this would be a mistake, which led to my decision to write the 1991 article. Nonetheless, within a few short months the Government succumbed to popular pressure and intervened to curb speculation. As described above, this was the first of many such interventions over the next 20 years.
There were major public concerns at the time about the methods used by developers to sell uncompleted housing units in advance and some of these concerns are still valid today. First, triad members wishing to turn a quick profit often used strong-arm tactics to jump the queue, making it difficult for law-abiding buyers to acquire the units directly from the developer. This concern is probably no longer important. Second, units sold by developers were marketed in ways that provided misleading information to the public. Third, final users who waited in queues on the announced date of public sale often found that many of the choice units had already been privately sold ahead of time to property agents or speculators through internal sales. Many buyers were therefore compelled to pay a considerable premium in order to acquire their units from the property agents and speculators; speculative activities were blamed for creating such a situation.
These concerns have a long history and each time property prices surge in the residential market, there are renewed calls for regulating the sales of these units so as to ensure fairness in the market and curb speculation. This article will not address the first two concerns, but deal with the more difficult and controversial issue concerning the need to curb speculation.
From an economic point of view, it is not obvious why speculation should be curbed. Speculation is economically desirable because it spreads risk. It can occur only when there are differences among people in their attitudes toward risk and their assessment of future prospects. So long as banks remain prudent in making mortgage loans, there is hardly any rationale for regulating a purely voluntary exchange arrangement. When the threat of systemic banking risks is absent, the case for market failure arising from speculation in asset markets cannot really exist.
Perhaps a case can be made that the pre-sale of new residential units in a large housing estate developments has an element of market power because of location-specific monopoly power, in which the developer has a dominant market position. It can be argued that developers of large housing developments estates may possess location-specific monopoly power, which is enhanced by the fact that housing units are not homogeneous making it legitimate to charge different prices for different units.
The location-specific monopoly argument cannot be ruled out in principle, but is it empirically plausible? To exercise such monopoly power, developers have to possess a mechanism that would allow them to sell units to their respective highest bidders. In other words, price discrimination must be present. This can only be achieved through auctions because developers cannot possess information as to who is the highest bidder for each unit in advance.
Location-Specific Monopoly Powers May be Spurious
And yet developers have chosen not to use auctions to sell their units, which is prima facie evidence that exercising price discrimination to maximize profit cannot be the explanation. Moreover, according to economic theory, price discrimination only allows developers to cream off all the consumer surplus of the buyers. It does not strictly speaking result in any economic efficiency loss that is the correct understanding of what market failure means.
It has been argued that auctions have not been adopted by developers because they are costly to administer, but it is not immediately obvious that there is much empirical merit in this claim. Even if this were the case, the monopoly argument still breaks down because it implies that price discrimination is too costly to implement. Since monopoly power cannot be effectively exercised, there is no case for regulation to constrain location-specific monopoly powers.
It is interesting to note that if auctions were adopted, short-term housing speculation would be reduced because the units would then be held by those with the highest bid. These are more likely to be final users than property agents and speculators because a competitive bidding process is likely to reduce significantly the scope for making speculative profits. This means that if price discrimination could be effectively employed, then the incentive to engage in speculation would be reduced significantly. In other words, speculation takes place only when monopoly power, if it exists at all, is not exercised.
Location-specific monopoly power, however, is not the same as monopoly power over the whole market. The latter involves an alleged ability to reduce the total supply of housing units in order to raise prices. WhileHong Kong’s property development market is heavily concentrated today, it is not obvious that developers can control the total supply behaving as if they are in collusion with one other. It is a common mistake to use monopoly power to explain price increases. Prices will rise if a competitive market is monopolized, but this is only a one-off effect. It can only explain why prices are higher, not why prices continue to increase.
If prices were to rise continuously over time, as they do inHong Kong, it would have to mean that monopoly power is also continuously rising. For this to make sense it is necessary to assume that not only is there monopolistic collusion among developers, but their monopoly powers have been continuously increasing. This cannot be correct or sensible theoretically or empirically. A monopoly power argument cannot account for the large price surges that we have seen in the past 20 years.
If prices were to rise continuously over time, as they do inHong Kong, it would have to mean that monopoly power is also continuously rising? For this to make sense it is necessary to assume not only there is monopolistic collusion among developers, but that their monopoly powers have been continuously increasing. This cannot be correct or sensible theoretically or empirically. A monopoly power argument cannot account for the large prices surges that we have seen in the past 20 years.
The puzzling issue that has to be explained is why do developers choose to sell their units the way they do?
One explanation is that while developers can make a reasonably accurate forecast of the average market price, it is impossible for them to know the highest bid for every unit. If developers are less efficient at searching out the highest bid for every unit, then one would postulate that speculators and property agents may be more efficient at this type of search activity to perform such a function.
Such an explanation, however, fails to explain why developers are keen to sell many of their units to speculators and property agents in advance. Since developers cannot gain from a better matching of units to final users, why do they care whether the units are sold to speculators, property agents, or final users? The explanation is also unconvincing because unless auctions are too costly to administer, developers will have an incentive to use them to force buyers to reveal their bids for every unit and capture the gains that would otherwise accrue to speculators and property agents.
In my next article I shall explain this and also consider the consequences of repeated attempts to curb speculation on housing prices in the market.