Land Auctions

Stephen Ching

M5, 2023-24

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Utilization

Hong Kong’s total land area amounted to 110,800 hectares as at

end-2011, of which about 23.9% or 26,500 hectares were developed
land. Non-built-up land accounted for the remaining 76.1% or
84,300 hectares of the total land area as shown in Table 1.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Utilization

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Utilization

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Supply Strategy

Hong Kong’s property market had undergone a distinct downturn

in the aftermath of the Asian financial crisis in 1997 and the
Government subsequently shelved the plan of increasing land
supply to meet the target of building at least 85,000 flats a year.

Since then, market forces have been employed as the principal

strategy for land supply management, with minimum government
intervention to ensure an effective adjustment according to the
supply and demand situation.

As such, sale of government land is subject to competitive bidding.

Sale price is determined entirely by market demand and reflects
what property developers are prepared to pay under prevailing
market conditions.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Supply Strategy

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

Before 1999, the disposal of government land was primarily

governed by the annual Land Sale Programme comprising a
timetable of specific sites to be put up for scheduled auctions and
tenders.

In 1999, the Government supplemented the Land Sale Programme

with a market-led mechanism, the Application List system
(ALS), allowing property developers to decide the timing and
quantity of land to be put on sale in a flexible manner so that the
market would not be overloaded with excess land supply.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

On 13 November 2002, the Government announced a package of

measures to restore public confidence in the property market,
which had been on the decline since the outbreak of the Asian
financial crisis. These measures included, among other things,
stopping all scheduled land auctions and suspending the ALS until
the end of 2003.

After the moratorium, the supply of new land could only come
from the ALS.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

On 15 October 2003, amid initial signs of a rebound in the

property market, the Government decided to resume the ALS from
January 2004.

The ALS had since then been the sole mechanism for the disposal
of government land until 2010, when the Government fine-tuned
the land supply arrangements by initiating land auction or tender
in a limited way with a view to stabilizing flat prices.

For selected land sale sites, the Government specified in the land
sale conditions the minimum number of units and/or unit size
restrictions for the purpose of increasing the supply of small and
medium-sized flats.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

In the light of continued surge in property prices and the limited

impact of the ALS on boosting the housing land supply, a more
proactive approach was taken by the Government in the
2011-2012 Land Sale Programme. Apart from the Application
List, the Government put up additional sites for disposal by tender
on its own initiative.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Annual Land Sale Programme

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Land Auctions

Is there any relevant theory to assess the decision to fully resume

the sale of sites by tenders only?

We can start with the standard comparison between (open)

auctions and (sealed-bid) tenders.

We assume that a land auction is a common value auction, i.e.

the value of the site is common to all bidders. However, the
common value is unknown to the bidders.

Each bidder has a private estimate of the common value. The

private estimates are assumed to be drawn independently from a
distribution with the mean equals to the common value, i.e. each
private estimate is an unbiased estimate of the common value.

Does it imply that the private estimates have the same value?
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Bidding Strategy in a Tender

Assuming that there are a large number of bidders, i.e. bidding is

very competitive.

How should a rational bidder bid in a (sealed-bid) tender? Does

she know the private estimates of the other bidders? Does she
know her own private estimate?

As a rational bidder, she should be able to figure out that her

private estimate is an unbiased estimate of the common value.

Will she submit a bid higher than her private estimate? Equals?
Lower?

Note that the tender is very competitive. Should her bid be lower
than, and very close to, her private estimate of the common value?
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Bidding Strategy in a Tender

Who will be the winner if all bidders follow the same bidding
strategy?

Will such a winner’s profit be positive? Negative? (Heard of the

“winner’s curse”?)

Is the above bidding strategy rational? How should a rational

bidder bid? (Heard of ”bid shading”?)

By using bid shading, rational bidders can avoid the winner’s curse.

What is the implication on the seller’s revenue? Should it be called

the winner’s curse or the “seller’s curse”?

It may not be a good idea to use tenders.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Bidding Strategy in an Auction

Does a bidder in an (open) auction observe the bids of other

bidders? Are the bids of other bidders related to their private
estimates?

Bidding of other bidders inevitably reveals some information of

their private estimates, which will be picked up by a rational
bidder. The additional information allows a rational bidder to
estimate the common value more accurately and, hence, rely less
on bid shading to avoid the winner’s curse.

What is the implication on the seller’s revenue? Is it a good idea

to replace tenders with auctions?

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Are Land Auctions Competitive in Hong Kong?
How can one assess competitiveness of a market?

Let’s start with a monopoly. When there is no competition, a

monopoly can make an economic profit. However, if the market
opens up, the incumbent’s profit will decrease, and in the extreme
case of perfect competition, will go all the way to zero.

Direct competition can be assessed by the number of firms in a

market and market concentration, but potential competition can
be as effective as direct competition. (Heard of contestability
(Baumol)?)

Economic profit provides a more robust assessment of

competitiveness/contestability of a market, but how can one
measure economic profit?

In general, it’s very difficult to measure economic profit, because it

involves measuring opportunity cost that is unobservable to
researchers. .
.
.
.
.
. . . . .
. . . .
. . . .
. . . .
. . . .
. . . . .
.
.
.
.
.
.
.
.
.

Stephen Ching Land Auctions

Are Land Auctions Competitive in Hong Kong?

We “outsourced” this difficult job to the stock market, believing

that the stock market is efficient, i.e. all publicly available
information is reflected efficiently on the stock prices (semi-strong
form).

In a land auction, if the winning price is “too low”, the winning

developer is expected to make an economic profit. If the winning
developer happens to be a listed company (many developers are
listed companies), the Efficient Market Hypotheis (EMH)
implies that the expected economic profit will be reflected as a
positive “abnormal return” in its stock price (after adjusting for
the “normal return” to the stock market).

Abnormal returns, as a measure of economic profits, can be used

to assess competitiveness/contestability of land auctions.
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Contestability of Hong Kong Land Auctions (1986-98)

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Contestability of Hong Kong Land Auctions (1986-98)

Applying the analysis to the land acquisition events at the

government land auctions in Hong Kong, we find positive expected
abnormal returns. The finding indicates that the urban land
market is not always contestable.

Furthermore, we find that the ex ante economic profit of land

acquisition tends to be higher when the auctioned site is more
expensive, the market liquidity is lower and the government land
disposal intensity is higher. These findings suggest that the
government can enhance the contestability of the local land market
by expanding the access to land credit, maintaining the property
market liquidity, and making the supply of developable sites more
elastic.

. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

Stephen Ching Land Auctions

Application List System (ALS)
The ALS is an attempt by the Government to make the supply of
developable sites more elastic.
The ALS is a hybrid system with two stages. The first stage is a
tender, requiring developers to submit sealed bids for the purpose
of releasing a site for an auction. An auction is triggered when the
Government receives a (sealed) bid from a developer that meets its
(undisclosed) minimum price.
The triggering developer commits to buy the site at the bid
accepted by the Government, should the subsequent auction fails
to generate a more satisfactory price.
Once a site is triggered, the Government will schedule and hold an
auction for the site, which is open to all developers.
Due to the winner’s curse in the first-stage tender, developers
shaded their (sealed) bids and failed to meet the (undisclosed)
minimum price set by the Government. .
.
.
.
.
. . . . .
. . . .
. . . .
. . . .
. . . .
. . . . .
.
.
.
.
.
.
.
.
.

Stephen Ching Land Auctions

Keypoints
It took the Government a while to realize that the ALS had
failed
Prior to that, the Government misunderstood that the demand
for residential land (and private housing) had been low
Consequently, the supply of residential land (and private
housing) had been insufficient
The Government should have fixed the ALS, instead of
replacing it with tenders
Tenders expose the bidders to the winner’s curse
Bid shading can be used by rational bidders to avoid the
winner’s curse and turn it into the seller’s curse
The bidders in auctions are less vulnerable to the winner’s
curse
Auctions are preferred to tenders, revenue-wise
The Government should consider to fix the ALS, because it
has potential to make the supply of (residential) land more
elastic and land auctions more contestable .
.
.
.
.
. . . . .
. . . .
. . . .
. . . .
. . . .
. . . . .
.
.
.
.
.
.
.
.
.

Stephen Ching Land Auctions