Ernst’s Confessions: The housing prices in The Netherlands are not so overpriced currently as I thought. But I doubt if that will set the housing market free in the coming years

Sometimes I get caught in one of those fierce Twitter discussions on the Dutch housing market. The question on hand is whether it is overpriced or not.

In the past few years, I looked at the price development in the period 1995 – 2007, when housing prices almost tripled in value: from 100 in 1995 to 270 in 2007 and I jumped to the conclusion that a real housing price bubble had developed in The Netherlands.

My stance until now had been that the Dutch housing market was about 30% too expensive in 2009 and as the housing prices had dropped about 10% since then, the housing prices were still 20% overpriced at the end of 2012.

Today I decided to look at things from another point of view. 

I took all the data from the period 1975 – 2011 that I could get. The used data is courtesy of the Central Bureau of Statistics (www.cbs.nl), the University of Amsterdam (UvA) and www.gemiddeld-inkomen.nl (i.e. average income) : 

• Average Mortgage Interest Rates (adjusted for average Mortgage Interest Deduction (MID)) 
• Official Inflation rates (by the Central Bureau of Statistics) 
• Average Housing Sales Price per year (UvA and CBS)
• Average Income (www.gemiddeld-inkomen.nl)
• Housing sales numbers (CBS|only since 1995)

I started with the average housing price in 1975 and I divided it by the average income, multiplied with a factor based on the deviation of the at-the-time interest rate from the highest interest rate over the whole period (adjusted for mortgage interest deduction).

I used this interest factor, as in times with low interest rates people can borrow more money than in times with high interest rates, based on the same income.

The formula looks something like this:

X = Housing Price 1975 / Average interest rate adjusted income 1975 (Average Income 1975 x (highest MID interest rate [1975-2011] / interest rate 1975)).

Then I divided the average income and the Average housing price [1975-2011] through the aggregated inflation rate to get an inflation adjusted income and an inflation adjusted housing price.

Using the earlier calculated multiplier X, I calculated the realistic housing price with a factor for every year from 1975 - 2011, based on the thesis that people in 1975 paid a price for a house that matched to their income. 


Hence, the multiplier X represented a reasonable proportion between the housing price and the income, adjusted for the interest rate at the time.

Average interest rate adjusted income ^[year] x [multiplier X] = realistic housing price ^[year]

Then I matched the real average housing price in a year with the realistic housing price that I calculated based on inflation corrected income, interest rate and the multiplier X.

I must admit that I don’t know if this method is scientific, as I’m not a econometrics scientist.

However, for me the method seems reasonable and defensible.

Here are some preliminary conclusions:

• The inflation adjusted income rose by only €470 in the period 1975 – 2011 (see following chart). 

Inflation adjusted income 1975 - 2011
Data courtesy of : www.cbs.nl and www.gemiddeld-inkomen.nl
Chart by : ernstseconomyforyou.blogspot.com
Click to enlarge

• There is a negative correlation between the inflation-adjusted housing price and interest rate. This works much stronger than I expected. I dare to say that scarcity on the housing market doesn’t play so much of a role.
  
• Although the inflation adjusted housing price in 2011 is about twice as high as in 1975, this can be totally accounted for by the much lower interest rates of today. In 1975 the average interest rate on a mortgage in The Netherlands was about 9%, today it is about 4.5%. When the inflation is pulled out of the equasion, the influence of the interest rates on housing prices is enormous (see following chart)

Inflation adjusted housing prices 1975 - 2011
Data courtesy of : www.cbs.nl and the University of Amsterdam
Chart by : ernstseconomyforyou.blogspot.com 
Click to enlarge

• However, it takes the housing market a few years to react to adjusted interest rates, especially when a recession or a crisis just took place before the interest adjustment.
  
• The Mortgage Interest Deduction dampens the effect of higher and lower interest rates, as the interest rate is about divided by two.
  
• After a crisis took place, the realistic housing prices are often much higher than the real, inflation adjusted housing prices. This situation can last for five to seven years (see following chart)

Inflation adjusted housing prices  vs realistic housing prices 1975 - 2011
Data courtesy of : www.cbs.nl, www.gemiddeld-inkomen.nl and the University of Amsterdam
Chart by : ernstseconomyforyou.blogspot.com 
Click to enlarge

Main conclusions:

• In the period from 2006 – 2010 the housing market was overpriced.
  
• From 2007 – 2009, the housing price was massively overpriced (25% in 2008). This explains why the housing market sales collapsed in 2007 and didn’t pick-up until 2010 (see the sales line in the last chart).
  
• At this moment the housing market is not overpriced anymore. The real inflation-adjusted housing price is exactly equal to the realistic housing price, if my calculation method is correct
  
• However, I expect the real housing price to be below the realistic housing price for at least 5 - 8 years. This happened too in earlier crises. Besides that, this was an extraordinary crisis with probably longlasting side-effects.
  
• On the other hand, I don’t expect the housing prices to drop much further than about 5 - 10% from today’s prices, as they seem to be the right price. The deep crisis, however, and the rising unemployment makes this very hard to predict.

I must admit that I am truly puzzled by the results of my research. Therefore I called this article 'Ernst’s confessions'.

One of the biggest scientific errors that can happen is that people try to adjust the data to their convictions, instead of adjusting their convictions to the results of the data.

Today, I try to do the latter, by admitting that my stance on the overpricedness of the Dutch housing market was correct in 2008 - 2009, but is now not correct anymore. I learned an important lesson today.