Housing shortage: When is a shortage not a shortage

By Matt Nolan 05/05/2014


It is received wisdom that Auckland has had a housing “shortage” for a prolonged period of time due to insufficient supply.  However, on the face of it this is a bit strange as:

  1. Residential investment:  Yes building rates have been low during the past 7 years, but the value of residential investment in the prior 7 years was very high.  The “volume” was lower, but what does that mean (we’ll come back to it).
  2. Growth in rents have been low.  Low low low.

In this way many people have, justifiably, questioned the idea of a shortage.

However, there is one potential explanation that could match both these facts and still give us a “shortage” that can help to drive the high price for housing – a fundamental imbalance in the supply of “types” of housing.

Often you will hear economists say that renting a house provides a housing service, and buying a house is just purchasing a stream of future housing services – therefore the house price must represent future rents.  This is true.  However, the “service” provided by a rented house and a purchased house need not be the same thing!

As a base way of thinking about it, a service from housing may have two components – a “need” component (eg I need somewhere to sleep at night) and a “want” component (eg I want a house with a large lounge for my tupperware parties).  Combined with the fact that renting and owning have fundamentally different flexibilities (when you rent you have an option to buy a house, when you’ve purchased a house you rely on the liquidity of the housing market to buy a new house), and different social views on renting and buying, the rental stock and the owner-occupied stock exist to satisfy different bundles of housing services.

In this way, the rental stock, which people are only interested in renting (except investors) is only sufficient to meet the “needs” base of housing, while the owner-occupied stock meets some of the “want” service of housing.

In this way, yes both types of housing are substitutes for the needs part of the service, but in terms of these “wants” there is no substitute for owner-occupied housing.

As a result, the “shortage” in this context could be just a shortage of housing for owner-occupied purpose – while the low rent growth is indicative of a potential oversupply in rental property.

Now this argument isn’t particularly complicated, and I think it meshes with the views of many – but it is interesting to think of what it implies:

  1. Is this really a “problem”?  If needs are being met by the housing stock, and people are bidding up prices with regards to wants, the case for “social injustice” from housing is significantly weaker!
  2. What social expectations do people have if they are so willing to bid up prices based on these “wants” for the housing stock – what are these wants and what do they mean? (eg unrealistic expectations, status goods, a product of falling material scarcity)
  3. Why has this occurred (if it has)?  Is it the structure of investment?  Is it investors confusing the average return on housing with the return on rental stock (thereby “accidentally” accepting very low yields on housing)?  Is it due to an institutional or policy related bias?
  4. Will the housing accord “help”?  If the issue is the type of housing, and there is some bias, does making more land – or in Labour’s case building lower quality housing – address the issue?  If this is the “problem” and we decide it is worthy of policy, then aren’t the policies of both political parties beside the point?

Now I’m not saying this is the way to consider it.  However, I’d note that if this is your narrative for the housing shortage it leaves a lot of important unanswered questions – and on the face of it shows that the run up in house prices (with banks holding sufficient equity) is a lot less policy relevant (as it doesn’t impact upon the minimum bundle of goods and services available to the worst off)!

A big criticism of this theory is that I don’t think it really matches the data.  If we were to use house size as a proxy for quality, additions to the housing stock got a lot larger between 2000 and 2008 – and have levelled off now.  This is a pretty big empirical shortcoming of this explanation!  However, it is still a useful thought experiment – and who knows, if we go through the data in detail we may be able to explain this, and we may find this is a good description of what happened, so considering it can’t hurt ;)