Nuggets 9 – How best to measure the housing market?
The origins of the metre go back to at least the 18th Century. There followed many attempts to standardise it as a unit of measurement. It took until 1983 for the metre to be precisely defined as the length of the path travelled by light in a vacuum during a time interval of 1/299,792,458 of a second. We could all sleep safely in our beds knowing that this problem had been so completely solved.
The reason that this was worth taking two centuries of experiment and debate to get right is that when examining anything, we need to be able to measure it fairly accurately and consistently, otherwise we have no objective sense of perspective or comparison.
Throughout Scotland, where we used to variously measure land in units such as ‘roods’, ‘groatlands’, ‘oxgangs’ and ‘pennylands’, a proper imperial and then metric system unquestionably made life easier for those dealing in land and property.
However, with house price measurement, analysis and reporting still seems to thrive in a realm of inaccuracy and inconsistency. Look at these 2 articles ‘Scottish house prices see further slide’ and ‘slight rise for Scottish housing market’ published within one week of each other. They each give very contrasting pictures of how the Scottish housing market is performing over the same time period. According to these, things are either going very badly, or not changing much at all. As a house buyer or a seller, developer or investor, what are you to make of all of this? One of the reasons that the housing market has failed to recover is that people cannot price assets properly and, with all of the contradictory information around, who can blame them?
The problem with house price indices
Most commentators seem to believe that house prices are the best way to measure the performance of the housing market. However, there are many house price indices. The assessment of house price movements in recent times from all of the leading indices can be best represented by a graph from LSL/Academetrics, which produces its own index.
Figure 1: Annual house price change in Scotland, 2012-13, various indices
The indices are measuring different things in different ways, therefore it is no real surprise that there is a lack of consistency. However, this only breeds confusion. We really need most of these indices to lay down their lives for the cause of clarity and consistency, but, as they all think they are right, this type of self-sacrifice seems unlikely.
To briefly summarise all of these.
- The Nationwide and Halifax indices are up-to-date, producing results quickly for the quarter just past. They also ‘smooth’ the data to take account of seasonal fluctuations and different mixes of properties sold. However, these are indices of agreed prices (Nationwide) or mortgage offer prices (Halifax) of properties that they have sold, therefore they are not complete indices in any sense – they only each cover around one-tenth of the mortgage market.
- Registers of Scotland does include everything bought and sold, but the time elapsed between missives concluding and registration is 2-3 months on average, therefore is not up-to-date. It also does not adjust its figures for seasonal factors (seasonal adjustment) or building types (mix adjustment), therefore is more volatile.
- LSL/Academetrics does take these factors into account in ‘smoothing’ the official Registers data. But ‘smoothing’ is not an exact or precise science itself. The Registers data is also full of ‘noise’, or, as others may more unkindly put it, ‘gaps’ and ‘errors’. Using it to provide a definitive house price index is not that easy and its results based on it should be treated with some caution.
- CLG/ONS data are based on a sample of mortgage completions prices and are mix adjusted, but not seasonally adjusted.
And that is not all of them! Lloyds TSB has its own price index for Scotland and there is the monthly sentiment survey by the Royal Institution of Chartered Surveyors (RICS) as well.
A far better way of looking at things
Given all of this, you might want to take future housing market information with more than a pinch of salt, but, in our view, there is a far better (and probably simpler) way of viewing the market that relies on all property sales and does not require training in hedonic modelling techniques.
As Oscar Wilde once famously said, “a cynic knows the price of everything and the value of nothing.” And there is a certain cynicism in all of the coverage of house prices – designed very much to play on peoples’ fears and greed – rather than provide a genuine and honest assessment of where the housing market is and could be going.
It is surely better to take a leaf out of Oscar’s book and instead look at value. The market value (or turnover) of the housing market looks at the total value of everything sold in the market in a given time. It therefore combines both prices and transaction levels. If you look at it on the basis of annual change, there is no need for seasonal adjustment, and there is no need for mix adjustment either, as the objective is not to produce a reasonable price series but only look at what the market is turning over.
The average price series (regardless of which one you look at) tells us little of what has actually been going on in the housing market over the period 2006-13. If after a 7-year visit to Mars, and desperate for news on how the Scottish housing market was performing, you looked at the average price series, you would be forgiven for thinking that the market had merely consolidated and stabilised in this time.
Figure 2 Scottish average house prices, 2006-12
Source: Registers of Scotland
Of course, the market has instead been through a tumultuous and unprecedented boom and bust, before only recently stabilising at much lower levels of activity. This is, in fact, the story that an assessment of market values gives you.
Figure 3 Value of residential property sold in Scotland, 2006-12
Source: Registers of Scotland
The real story of the housing market is that it rose rapidly in 2001-07, before crashing from late 2007 to early 2009. It then stabilised and it has been slowly recovering in the last 3 years. However, it is still around half the levels of its peak. It will recover in time, but it is likely to be protracted.
So, now you have the measure. Whether the glass is half full or half empty though is a matter for your own interpretation.