A brief guide to real estate barometers
With the start of a new year, we all like to look back on the previous year. The real estate market is no exception, and Realo is no stranger to this practice (you can still read their annual report here). This means that in early January, various real estate barometers appear, which at first glance may seem to contradict each other. However, nothing could be further from the truth, although you do need a good understanding of the underlying data and methods to see the wood for the trees.
Asking prices versus selling prices
First and foremost, it is important to understand exactly what type of prices is being discussed. Anyone who has ever ventured into the housing market knows that the asking price is not always equal to the selling price. Both figures are at two ends of an elongated process: from the moment an ad is placed online to negotiating the price and finally signing the deed and handing over the keys. Such a process can easily take four to six months.
Conclusions based on selling prices are somewhat behind those based on asking prices. Only when the deed is signed do the selling prices enter the statistics. However, in the meantime, the asking price has already been included in the asking price statistics six months earlier.
What are the implications? A concrete example: after the pandemic placed all of Belgium, including the real estate market, under lockdown in 2020, buyers returned to the market at the beginning of 2021 and were eager to purchase something (larger). This high demand likely led to an overheated real estate market in the first half of 2021, resulting in strong price increases. In other words, at the beginning of 2021, asking prices started to rise sharply. Six months later, when the properties had definitively changed ownership, the final selling prices also entered the statistics. However, this was only six months after the overheating had begun.
Are asking prices more interesting than selling prices? Not necessarily, it depends on what you're looking for. Generally, the selling price is still slightly lower than the original asking price. So, if you're interested in how much buyers paid for a property, you're better off with the selling price - with the caveat that the situation six months ago may not be the same as today. If you want to know how prices are evolving in your neighbourhood and what is being asked for a property today, then you're better off with asking prices.
Houses versus apartments
One property is not the same as another. Are we talking about houses or apartments? They can experience a very different price evolution depending on their attractiveness. Now that energy consumption is higher on buyers' priority lists, the choice between a house or an apartment may matter more. Apartments generally have lower energy consumption due to their enclosed nature, which may make them more attractive to buyers compared to a year ago. Lump-summing houses and apartments together can therefore give a distorted picture.
Comparing cities can also lead to incorrect conclusions without making a correct distinction. In Brussels, the market consists of approximately 80% apartments. Compare this to a city like Ghent, where apartments make up only 40% of the market, and you can see that a fair comparison becomes very difficult.
For these reasons, at Realo, we make a distinction between houses and apartments. At Statbel, for example, the statistical bureau of Belgium, they go a step further and distinguish between houses with semi-detached or detached construction and houses with open construction. So, always look carefully at which type of property is being discussed and what distinctions are being made.
Averages versus price models
Not only can the data used lead to different conclusions, but also how you handle that data can be decisive for your conclusions. Some barometers will work with average prices, while others (including Realo) work with a price model.
Again, a (simplified) example. Suppose you live in a village with five houses and a castle. Last quarter, two houses were sold, averaging €250,000. The next quarter, another house and the castle are sold. Suddenly, the average selling price of your village has exploded to €850,000?! Is your village suddenly the most sought-after place in Belgium? Unfortunately not, the castle is so much more expensive than the rest of the village that the average price was unrealistically inflated. Of course, this is a simplified example, and often municipalities with very few sales will not be included in the statistics because they are not reliable enough. But even for larger cities, a price increase between quarters can sometimes be explained by the type of properties that were sold. For this reason, Statbel, for example, will report median prices, a statistic that is less sensitive to exceptional sales.
To counteract such a distorted image, pricing models are sometimes developed. In the annual report of Realo, for example, we work with such a model. The advantage of a pricing model is that it explicitly takes into account the characteristics of a property. An extra bedroom will receive a certain surcharge through the model. If there are fewer houses with 3 bedrooms instead of 2 bedrooms on the market in a particular quarter, this is no problem for the model, as the added value for a house with more bedrooms is independent of how prices are evolving.
The question arises again whether one is better than the other. And once again the answer is: it depends. Pricing models allow for direct comparisons between municipalities because you can assume a 'standard' house or apartment with the same characteristics, but placed in a different municipality. In other words, you are not comparing apples to oranges. The disadvantage is that the chosen 'standard house' may not be representative of your municipality at all. In that case, it may be more interesting to look at average prices.
If you want to look at price trends over time, then a pricing model is better. Such a model can take into account the characteristics of the properties that have been sold and therefore works independently of the characteristics of properties that happen to have been sold in a particular quarter.
Conclusion
To summarise, no real estate barometer is wrong. Much depends on which data is used and how that data is handled. By being attentive to these points, you can already draw many conclusions. Ultimately, the most important thing is that you can draw the right conclusions for yourself that are most relevant to you. Or consult Realo, then you are sure.
