Building Costs

When someone is in the market for real estate, one can buy an already build house/apartment or one can buy a newly constructed house/apartment. In theory, one will always choose the cheapest option but as discussed previously it is very difficult to see the assets in the real estate market as similar. Often locations and building materials are unique which does not make real estate a commodity. Buyers are therefore forced to buy what is available in the market instead.

As mentioned in section 3.6.1, in the short run the supply is fixed as it takes time to develop and build new real estate. This is not true in the long run. In the long-term perspective one can use the ratio of building costs relative to the prices in the market to determine whether the market looks overheated for some periods.

Figure 16: The Building Costs vs. the Real Estate Prices from 1992 until 2020

Note: Own creation based on Danmarks Statistik’s databases BYG4, BYG42 & EJEN77

38 In Figure 16, building costs are presented based on quarterly observations from 1992 until 2019. The dataset is constructed as a chained dataset based on two different datasets. The first dataset is BYG4 from Danmarks Statistik which had observations from 1992 until 2015. The second dataset is BYG42 which has observations from 2003-2019. These two datasets were merged into one dataset going from 1992 until 2019 with a base year of 2010.

The building costs pattern looks a lot like the GDP growth and the disposable income development. They increased throughout the whole period with the biggest increase around the years of the financial crisis.

This was followed by a minor decrease in prices shortly after the financial crisis. From 2010 and onwards the price has shown a steady upwards trend with no major fluctuations.

Interestingly, the graph also shows that before the financial crisis it was beneficial to buy an already constructed house or apartment compared to the building costs. They were significantly higher than an equivalent house or apartment. However, from 2004 and until the start of the crisis the picture was the opposite. In that period, the prices of new homes were significantly cheaper compared to the market prices. This pattern is exactly the same as mentioned in section 3.6.1, where many more homes were built from 2004 until 2008 compared to the period from 1992 until 2004.

The price of buying an apartment today is significantly higher than the costs of building one. However, the price of an already constructed house compared to the building costs of a new house is relatively cheaper today. This indicates that the price of already built apartments is potentially too high. The shift in prices between building costs and market prices changed in 2015. This explains the increase in finished building projects that has increased every year from 2015 until 2019 as also were mentioned in section 3.6.1.

4.8.1 Tobin’s Q-Ratio

Tobin’s Q was presented and defined in section 3.6.1.1. It is useful to elaborate on it further as the supply in the real estate market is inelastic in the short run. The Tobin’s Q ratio is therefore a useful tool in order to take the temperature of the real estate market over a longer period. If the Tobin’s Q-ratio increases it indicates that demand is increasing in such a way that the supply cannot keep up. This can potentially lead to a bubble, which makes it interesting to outline.

In Figure 17, the Tobin’s Q-ratio is presented for the housing market and the apartment market in Denmark and Copenhagen. The ratios are calculated from the index data that was already presented in sections 4.8 and 4.1 with a base year of 2010.

One can see that the Tobin’s Q follows the overall pattern for most of the fundamentals quite nicely. It peaks around the financial crisis and when the bubble started to collapse in the financial crisis years, the

39 Tobin’s Q ratio converged down towards the structural level again. Since 2010, the ratio has increased again for all three defined markets.

Figure 17: Tobin’s Q Ratio from 1992 until 2020

Note: Own creation based on Danmarks Statistik’s databases BYG4, BYG42 & EJEN77

I have added two extra trend lines in Figure 17. They show the average growth rate of the housing market and the apartment market from 1992 until 2004. The trend lines help us to see the structural differences between the “normal” level and the times when the market was out of equality. The period from 1992-2004 was chosen as it had a fairly stable increase over the whole period with no significant fluctuation.

Hence, it serves as my proxy for a “normal market scenario”. However, it is very difficult to define such a scenario. Another time period could have been chosen as well.

It is interesting to see that the trend line estimates a higher ratio in 2019 than what actually happened. If the predicted trend line had turned out to be correct, there would have been a clear mismatch between prices and building costs. As this is not the case, we can assume that the market is working well as it can regulate itself. For example, if housing prices were too high, entrepreneurs would begin to build new real estate as it is relatively cheaper to do this compared to buying an existing property.

40