The global financial crisis of 2008 made it clear that insight into the emergence of real estate bubbles is of great importance. However, especially in commercial real estate, this is difficult to obtain. CBS statistician Farley Ishaak investigated how we can gain insight into various hidden transactions in this market. He defended his thesis on this subject at TU Delft on April 25, 2025. With more accurate statistics, we could have seen earlier that things were not going in the right direction, says Ishaak.
The trade in commercial real estate has a very different character than that in the residential market. Not only is the market more heterogeneous with objects that differ enormously in size, but these transactions often involve multiple objects as well. Instead of a single house, entire packages of real estate sometimes change hands. These are so-called portfolio sales where, for example, two commercial properties, ten shops, and twenty homes are sold together. Sometimes large numbers of rental properties also change hands all at once from one investor to another. If these commercial transactions also occur as share transfers, the contents are hidden from view.

Market and Price Developments

These share deals, as it is called in real estate jargon, obscure the picture and result in a low number of observations, explains Ishaak. No assets are being traded, but shares in empty BVs in which the real estate is housed.
And that is a problem when you want to keep track of market and price developments as a supervisor. It is unclear what prices are being paid per object in such a share transaction. In the transfer of large real estate packages, there is often a quantum discount, just like in the supermarket. If you do not know the amount of that discount, you do not know the actual prices paid.

Quantitative Research

Ishaak encountered the problem through his work at CBS. He studied public administration, but after his studies, he mainly focused on quantitative research: putting numbers in a row and comparing them. Since 2013 he has been doing this at CBS, as production leader for construction and real estate prices. This department publishes the price developments of Dutch buildings monthly. Since the financial crisis of 2008, after the bankruptcy of investment bank Lehman Brothers, his department was given a task by the International Monetary Fund (IMF) and the G-20: to help clarify real estate bubbles.

Insight into Commercial Real Estate

Like colleagues in other countries, the statisticians at CBS developed indicators to provide insight into the real estate market. How do you do that in commercial real estate? Usually, the methods used for the residential market are applied without adjustments. But that results in a large margin of uncertainty. Share transactions in this sector are not always transparent. Ishaak calculated that more than 30 percent of all transactions in the Dutch real estate market are hidden in share deals. Portfolio sales also occur in large numbers. The method devised by the PhD candidate begins with splitting portfolios. He adds up the WOZ values of offices, commercial properties, homes, and shops. These values have been requested from the Cadastre. Based on that, he creates a distribution key with the percentage that has been paid for each object in the total amount. In addition, he developed a new index method, aimed at low numbers and heterogeneous observations. Using the open-source programming language R, he added this calculation method to the R package cbsREPS. Because this package is included in the public repository of R (CRAN), the method can be applied worldwide by R users.

Housing Price Developments

Ishaak also used his method to calculate housing price developments at the municipal level. Existing methods were previously insufficiently accurate. Therefore, CBS and the Cadastre only published the price developments for the four largest municipalities. By using price information from previous periods, more accurate calculations are possible. The result is that housing price developments for 306 municipalities are now being published. The development of plot prices is also being calculated with it.

Insights

What insights do these improved techniques provide? Regarding housing prices: they have risen in almost all municipalities by the end of 2024. Edam–Volendam stands out because there, prices fell at the beginning of 2023 and did not show any recovery for the rest of the year, says Ishaak. In 2024 we will only see a recovery movement with a greater difference as a result. Are we actually heading towards another real estate bubble? It does not seem so, because the price development of both residential and commercial real estate flattened out a bit in 2023. In addition, you will need to look at the financing side of real estate for that. Because the loan amounts in the previous decade have been limited and incomes have risen, we seem to be better equipped against a real estate bubble than in 2008. 

Applying Knowledge

The new method devised by the PhD candidate has already proven its usefulness: CBS is already using it for calculating the real estate figures presented every quarter. Statisticians elsewhere in Europe are also exploring the method for use. Ishaak: Great, because knowledge should be applied.

Relevant Links