Sunday, 15 June 2014

How a Danish king ruined some hopeful research

This post on VoxEU purports to show that there is no house price bubble in the UK, and perhaps more importantly, that there was no house price bubble in the run up to the 2007 crash.

This may come as something of a surprise to those who are used to looking at charts like this:


This chart shows the rise in average UK house prices since 1973 (source: Nationwide). There is clearly a steep rise in prices from 1995 to the crash in Q4 2007, which dwarfs the previous housing boom and bust in 1987-8. Many people would regard such a dramatic rise and fall as evidence of a bubble.

But the VoxEU authors don't think it is. And they base their argument on a unique feature* of the UK housing market - the existence of very long-term leases as well as freeholds.

Typically, in the UK housing market, houses are freehold and apartments leasehold, though this is not universally the case. A lease is a right to use a property for a period of time. In effect, it is a time-limited, pre-paid form of ownership. Properties that have leases close to expiry decline in value rapidly, making them difficult to sell, so most leases are automatically renewed. So that leases don't have to be renewed very often, many are very long-term - up to 999 years.

As the authors explain, such a long lease means that there should be no difference in value at its expiry from that of a freehold property. Both, at current prices, should be zero:
....the leaseholder owns the property for 999 years. The freeholder owns the property not only for the first 999 years, but also for all periods after that date. Therefore, the difference in the price of the freehold and the leasehold reveals the value today of owning the property 999 years from now. Ownership of a house 999 years from now is essentially a claim to the bubble asset. It entitles the owner to a single payment (the value of the house) in 999 years. This is so far in the future that it should have no value today, even if discounted at extremely low rates (say 1% per year).
They then go on to show that there is indeed no difference in price between properties with leases of 700 to 1,000 years and freehold properties. And they conclude:
We find that extremely long leaseholds are valued identically to otherwise similar freeholds. Our results, therefore, show no evidence for infinitely-lived bubbles in these markets. We find this to be true even in geographic regions where people argued that housing bubbles were likely to be present (e.g., Prime Central London).....the price discount is small and not statistically different from zero...in every year since 1995. In particular, we find no evidence of infinitely-lived bubbles even during the house price boom that culminated in 2007
This is simply absurd. It is impossible to predict what will happen at the expiry of a 999-year lease. The property may not even exist any more, and nor might the laws that govern the lease. Just to put things in context, 999 years before the present date was the year 1015. That was prior to the signing of Magna Carta, which is the foundation of all present UK law. It was prior to the Norman Conquest. It was prior to the building of most of the surviving castles and cathedrals in the UK, let alone the houses. It was, in fact, during the Danish occupation of England, and the reigning king was Aethelred the Unready (he died the following year). If there was a system of law in England at that time, it was - presumably - Danish. Would documentation for a lease issued under Danish law in 1015 AD even exist today, let alone be recognised as legally binding? Who would own the lease, anyway? We don't have many Danish thegns left in the UK now.

For all practical purposes, a 999-year lease is indistinguishable from a freehold. The only reason for having leases of such length at all is that lessees don't have quite the same rights and responsibilities as freeholders: they usually pay "ground rent" to the lessor, who remains responsible for maintaining the fabric of the building and communal areas. The fact that the authors could show no significant price difference between long-lease properties and freeholds demonstrates that the pricing of ground rents is generally efficient and fair. But it says absolutely nothing at all about the existence or otherwise of bubbles in the pricing of freehold properties. We are, in short, no further forward.

Personally I am of the opinion that there was a bubble in house prices from 1995-2007, which only partly burst. It was initially prevented from deflating fully because of severe supply restrictions and the collapse of the construction industry, and further deflation is now actively being prevented by government and central bank intervention.

The chart above suggests that there is no new housing bubble, not even in London: the fact that house prices remain elevated is simply unfinished business from the 2007-8 crash. But houses remain significantly overvalued relative to incomes. Either incomes must rise significantly, or house prices must fall. The question is how this can be achieved without trashing the economy. I admit, I do not have an answer. There is no simple solution.

Related reading:

The British obsession with property - Pieria


* Well, not quite unique. Singapore has these too.


7 comments:

  1. But houses remain significantly overvalued relative to incomes. Either incomes must rise significantly, or house prices must fall.

    Why would this be true? Incomes have to bear a relation to rents, not to capital values of houses.

    ReplyDelete
    Replies
    1. Rents usually track house prices, because if a large divergence develops between the cost of borrowing to buy and the cost of renting people adjust their behaviour to bring the two back into line. So unless you were thinking of introducing rent controls, incomes have to rise and/or prices have to fall to make both buying and renting affordable.

      There is an anomalous divergence in central London at the moment because investors buying properties for short-term speculative gain are leaving them empty. But this is not happening elsewhere and we should not view the whole UK housing market purely from the perspective of central London.

      Delete
    2. Another possibility is presumably to limit lending. If mortgages were only permitted to be 3xsalery, then presumably that would allow rents to increase without significant constraint from the threat of people buying houses, since buying would just not (generally) be an option.
      Just using such a measure might moderate house prices, but leave most of the houses being bought to rent (since few can afford to buy them to live in).

      Delete
  2. My father worked in a Canadian bank in the early 90's in the area of mortgage lending. His view (and mine by the way) is that house prices are correlated to credit availability - specifically income multiples. If house prices go above that relationship, then it becomes a bubble scenario.

    But also - London is a clearly segmented market and an international destination (i.e. like Hong Kong, Singapore, New York, etc...). This affects the top end of the market. The rest of London (outside zone 1) seems to be linked to the salary multiples effect. Add in the fact that we are short decent housing stock, then this forces prices upwards so that it bumps around the salary multiples top end/over.

    The lack of housing stock also affects the BTL market - if there is enough stock, then rents decrease as well. At the moment, there is also a definite lack of decent rental stock in the London area, which also leads to high rent issues as well.

    Please remember, average house prices skip over the market segments - We need stats that split out the segments and gives us a lower level of detail. For London, this is a critical effect.

    Additionally, be careful where you read the stats from, the Land registry lags behind the Nationwide/Halifax stats due to how the data is generated (and what it excludes). This is especially important if you are reading the daft house price headlines in the newspapers...

    Personally, I would like to see the following policy decisions:

    For London, shift the green belt boundary to the edge of the M25 OR consider 5 level mansion blocks like in Paris (a modern interpretation).

    For outside London - we need a "knowledge economy" home outside London (say Newcastle/Hull) with decent long term tax breaks + decent transport links to London for finance/international transport. (and) Consider some decent transport links to the some of the depressed areas of the UK - get the links into the town/business centers so people can easily get out of the sink estates and into the areas where there is work. (an easy hit is to subsidize bus routes that actually run at commuter times - 11AM bus links only help people get to the benefits office...)

    ReplyDelete
  3. For those of us attracted by geophysics, uncertainty and chaos something has to happen. Given that interest rates have been held down for so long and by so much it can only be a matter of time before something big will happen, but where, how, when, why are all questions without immediate answers.

    ReplyDelete
  4. Regarding the graph


    The second rise may look greater than the first due to the linear nature of the graph, but the first is in fact the greater. 13000/50000 = 2.6 50000/10000 = 5

    ReplyDelete
  5. What happened to the Rule Against Perpetuities? It used to be a core element of English law.

    When it was around, a 999-year lease literally was a freehold.

    ReplyDelete