Singapore HDB sale exercise

INTRODUCTION
This is about the Bi-monthly sale of 4-room and above flats by the Housing and Development Board (HDB) of Singapore. The recent offer of 278 units attracted 10240 applications @ S$10 non-refundable fee each.

PROBLEMS
Who are these 10240 people who applied? How many of them are serious and ready in buying? How many of them are trying their luck? Just for S$10 and a few minutes to apply online, they get a chance to buy a subsidized unit from HDB. The chance of 278/10240 = 0.02715 is a good chance to them for a S$10 bet, and a very high potential of gaining 50 to 100K in 5 years time.

However, this is a very, very, very slim chance for serious and ready buyers. ... How is it so? For example, if 80% ~ 8000 are trying their luck, using 10000 in place of 10240 for easy calculation, the expected number of applications from these 8000 getting a queue number in the first 2000 is 8000/10000*2000 = 1600. More specifically, on the average, every 4 out of 5 is a trying luck application. Thus, only 278*1/5 ~56 are serious and ready buyers out of the assumed 20% ~ 2000 of them, i.e. for serious buyer, the actual chance is 56/10240 ~ 0.00547 << 0.02715. Also, by assuming a low successful selection rate of 50%, the first 556 in the queue would have selected all the 278 units.

This system leads to these results. Does HDB want this? I am not sure if they have any priority system so that it is not a purely by chance system. (See my follow-up article on this.)

SOLUTIONS
They are reviewing the system now. Hopefully they can start off with a clear goal and objective in mind for changes and new implementations, rather than a problem solving mind set.

MY VIEWS
I only have vague idea about the earlier systems where it is purely first come first serve basis, and when there are many luck-trying people before serious buyers, the latter has to wait many years for their turn. Probably it was with a problem solving approach, that they dissolves the queue periodically so everyone can have some chance to be a first comer for each sale offer. Yet, long queues started forming up before official announcement of sale killed that system. Then, this present system is implemented where everyone who applies in the first week will be randomly given a queue number, and those who apply thereafter will just continue to queue behind. I guess the assumption is that the number of applications is about the number of units being offered. Sadly this isn't the case this time.

I feel that a week time is way too long a period for online application. My speculations is that most, probably 90% of serious and ready buyers would have applied on the first day. As the news spread out to more people, there will be many luck-trying people submitting their applications. Furthermore, there was this news article reporting that almost 9900 applications on Friday. It is supposed to serve as a deterrent telling people that the chance is low. Yes, indeed, many serious and ready buyers might feel so. However, for the luck-trying people, they might have read it as "Yes, the chance seems low, but this low chance or high volume suggested this is a good bet!" Hence, over the weekend and Monday, there are still another additional 250 applications.

Although something similar to the COE system might help to filter away those luck-trying people, it still has its cons and it is unfair to serious buyers. More importantly, this would be a problem solving way unless the expected resulting scenario is what HDB wants.

Thus, the crucial question is, what exactly does HDB want? What are its objectives and goals? Hopefully whatever new policies and sale systems, will not be for the purpose of solving current situation, but will be designed to achieve their objectives and goals.