At this place we take a look at short analyses, new data and interesting time series on the housing market. Feedback and questions are welcome at or via @empirica_regio.


empirica Housing Market Forecast

Authors: Jan Grade, Dr. Reiner Braun

The updated housing market forecast on the demand for new construction is now available for flats in single and two-family houses as well as flats in multi-family houses . The data are available via the empirica regio market studio and are included in our housing market reports. Here we would like to briefly and concisely present selected results of the housing market forecast. A more detailed discussion of the results can be found in the current empirica-Paper No. 256 (available in Germany).1

Demand for New Apartments until 2030

Three scenarios are considered for the forecast. Depending on the scenario, this results in an average annual demand for new housing for the whole of Germany of between 311,000 and 321,000 new flats in 2020, and an estimated demand of between 269,000 and 282,000 new flats in 2021. The following figure shows the estimated demand for new construction between 2020 and 2030 in the middle "base" scenario for the whole of Germany. The forecast shows the estimated demand, which is derived from the demographic development. However, it does not represent a forecast of construction activity. The construction completions of around 293,000 dwellings already available for 2019 are slightly below the demand estimate for 2020, but above the estimated new construction demand for 2021. However, this simple comparison for the whole of Germany does not take into account the regionally different distribution of demand and construction activity. In fact, in purely quantitative terms, too little construction tends to be carried out in urban growth regions, and too much in shrinking, often rural, regions.

Source: empirica / empirica regio (own calculation; data basis: Statistische Ämter des Bundes und der Länder, Deutschland, 2018-2020, dl-de/by-2-0)

Spatial Distribution of the Demand for New Apartments

Demand for new construction varies from region to region. The forecast takes into account not only the absolute new demand resulting from the change in the number of households, but also the qualitative additional demand. This means that there is also demand for new construction where the number of households decreases or where there are vacancies. This applies in particular to the segment of detached and semi-detached houses. New construction demand for multi-family houses, on the other hand, is mainly concentrated in the larger centres and their surrounding areas.

Annual regional demand for new construction according to the base scenario for 2021-2025, differentiated between one and two-family houses as well as apartment buildings

Is there a Need to Catch up?

When housing is scarce, i.e. when there is too little construction, and housing is therefore becoming more and more expensive, many households choose alternative locations. The migration data of recent years show that the influx into the (wider) surrounding area of the larger centres has increased. For example, the university towns, which have been cheaper up to now, have benefited more than before from immigration by young people. Therefore, a lack of new construction in the past does not necessarily have to be "made up for". If the demand for new construction and construction activity were to be balanced in the future, no additional shortage would be created. However, if rents in the strained regions are to be eased, there would be the possibility of either building more new homes here than the forecast demand estimates, or shifting demand to regions with low shortages or vacancies.

Forecast Methodology

The empirica housing market forecast represents an estimate of the demand for new construction, i.e. the additional housing needed in the future due to increasing population/household numbers. The estimate is based on the ranges of the 14th Coordinated Population Forecast of the Federal Statistical Office2, assuming an average annual migration balance between 147,000 (scenario "negative") and 311,000 (scenario "trend"). As a result, the population trend gradually flattens out. In the "negative" scenario, the population maximum is reached in 2024, in the "trend" scenario after 2030. Possible changes in migration to and from Germany as a result of the corona crisis are not yet taken into account. Furthermore, changes in household size and age distribution, current vacancies, and subletting and second home rates are also taken into account for the forecast. In addition, the qualitative demand for new construction is estimated.

1. Dr. Reiner Braun (2020), Housing market forecast 2021/2022. regionalised forecast in three variants with outlook until 2030. empirica-Paper No. 256. [in German]
2. Federal Statistical Office (Destatis) (2020), 14th coordinated population projection - base 2018.


Housing Market and Corona

Author: Jan Grade

Initial Situation Before the Corona Crisis

In recent quarters, purchase prices in Germany have continued to rise, while rents in many core cities rose less strongly than before or even stagnated. In recent months, this has led to a widening spread between rental and purchase prices. Since the fourth quarter of 2019, the labour market has shown the first signs of fatigue with stagnating unemployment figures, some of which are rising slightly. There has been much debate as to whether a turnaround is imminent in the current housing market cycle. However, with the outbreak of the corona pandemic and a massive economic slump, the outlook has changed completely. With regard to the housing market, the main question is now whether and to what extent Covid-19 is leaving its mark on the current housing market data. To this end, we would like to take a brief look at selected indicators and make an initial rough assessment.

Economic Development in Germany

The current economic figures from March 2020 onwards clearly show the effects of the corona crisis. The production index for March 2020 shows a decline of -6.7% compared to March 2019, the volume index for new orders in the manufacturing industry even shows -11.9%. At the same time, unemployment figures in March 2020 were 1.5% higher than in the same month of the previous year. In April the production index decreased by -25.3% compared to the previous year and new orders decreased by -36.7%. Unemployment was 18.6% higher than in April last year. This represents an increase of around 415,000 unemployed persons compared to the same month last year and 308,000 unemployed persons more than one month before. The latest data for May show a further increase of around 169,000 unemployed compared to April. This represents an increase of 25.8% in May compared to April. Other economic data are also clearly negative. Retail sales, for example, were 6.5% lower in April 2020 than in April 2019.1 The short-term economic consequences are already clearly visible.

Production index and volume index for May 2020 not yet available at the beginning of June
Source: own representation based on Federal Statistical Office (Destatis), 2020 (production index, volume index) and
Statistics of the Federal Employment Agency (unemployment figures)

Development of Housing Offers

Various analyses of housing market advertisements show that there has been a decline in the number of advertised residential properties throughout Germany since the contact ban on 16 March 2020.2 We wanted to know more about it and have therefore examined the advertised rented and owner-occupied flats in German cities with more than 50,000 inhabitans (the evaluations for various cities can be found at For this purpose we have evaluated the offers and prices by the starting date of the advertisement. The following figure shows an example of the data for condominiums in Frankfurt/Main. The results for all cities show: A noticeable decline in the number of advertised rental and owner-occupied flats can be observed since 16 March. The price development is less clear; neither rents nor purchase prices for condominiums have changed significantly since mid-March. The partially observed price decreases are within the usual fluctuations and cannot be attributed to Covid-19 alone. The decline in offers can also be seen in the data for Frankfurt/Main presented here. The purchase prices even rose initially in March and then returned to the average level of the last few months. Our colleagues from empirica accordingly stated that the property market was simply frozen in the short term3.

empirica regio (basis: empirica-systeme Market Database)

Google Trends

The German keywords "Miete" (rent), "Eigentumswohnung" (condominium) and "Stundung" (deferral) from the Google Trends for Germany reflect the course of the Corona pandemic much more strongly than the data of advertised flats. At the beginning of March, search queries for "rent" and "condominium" initially declined, but then queries for "rent" rose significantly in the third week of March. At the same time, the trend for "deferral" also rose significantly. This week saw the start of the discussion about protection against dismissal and deferral of payment for tenants. On 23 March, the Federal Cabinet adopted the corresponding Draft law to mitigate the consequences of the COVID 19 pandemic (press release in German). Parallel to this development, however, interest in condominiums seems to have declined significantly.

What is most interesting, however, is that although the trends deviated significantly upwards and downwards from the usual fluctuations (e.g. Christmas/New Year's Eve) in mid-March, they had already settled back to pre-Corona levels by mid-April. During this period, mortgage contracts for private customers were also increasingly granted again, while the number of new loan contracts stagnated at a low level.4

Interest in Germany according to Google Trends (Index 0-100, 100 = maximum popularity in the period)
Source: own evaluation based on Google Trends

Some housing market experts already suspect that in the future the home will even become more important as the centre of life, but also as a workplace (home office). Features such as flexible living spaces, access to the garden or a balcony will then gain in importance. The Google trends show that the keywords "garden" and "balcony" were much more in demand this spring in Germany than last year and the increased interest is continuing.5


The data clearly show the impact of the corona pandemic on the economy. The housing market, on the other hand, is reacting sluggishly and, measured by advertised flats, has so far only shown a decline in the number of advertises, but no collapse in prices. Whether and to what extent Covid-19 still leaves a footprint on the housing market will only become apparent in the coming months and will depend to a large extent on the further economic development. When new data on this will become available, we will discuss them here.

1. Federal Statistical Office (Destatis) (2020), Press release No. 192 of 29 May 2020.
2. VALUE AG (2020), VALUE Data Insights 1-2020. [in German]
3. Dr. Reiner Braun, Prof. Dr. Harald Simons (2020), Corona and property prices. empirica-Paper No. 255. [in German]
4. Latest data on residential mortgage contracts and retail loan contracts in Germany: Federal Statistical Office (Destatis) (2020), Lending and information for online transactions.
5. Google trends on the terms "rent", "condominium", "balcony", "garden" [in German].


Price Development of Curbed Apartments

Author: Dr. Lorenz Thomschke, translated by Jan Grade


In 2015 the so-called rent brake was introduced in some federal states in Germany. The aim of the initiative was to curb the price increase of newly contracted rents in prospering cities so that housing in these cities remains affordable even for middle and low-income households. At least that is how the draft law on the corresponding amendment to the tenancy law should be understood. ZEIT ONLINE has asked empirica regio to investigate the price development of newly advertised rental flats since the introduction of the rent brake in selected cities. However, the following evaluation is not a comprehensive study of the effects of the rent brake, but only an ad-hoc evaluation of the supply rents of individual cities.


Since the introduction of the rent brake, newly agreed rents in regions with tight housing markets may only exceed the local comparative rent by a maximum of 10%. This does not apply to comprehensively modernised flats and those that were first occupied after 1 October 2014. Flats that are rented for a limited period of time are not affected by the measure either. The new contractual rentals of these uncontrolled flats can still be freely negotiated. In contrast, the following evaluation is limited to the "braked" or curbed flats.1 In addition, the period under investigation was standardised in such a way that for each city there are six half-years before the respective introduction of the rent brake and six half-years afterwards.

Source: calculations by Lorenz Thomschke, empirica regio (basis: empirica-systeme Market Database)


The development of the advertised rents examined in the study suggests that there was no serious impact of the reform on price developments. At best, there is a moderate decline in rents in most cities between the six months before and after the introduction of the rent brake. However, the evaluation is a purely descriptive analysis which does not allow any direct conclusions to be drawn about the causal effects of the reform. For a causal-analytical evaluation, one would have to be able to observe the affected housing in a world where the rent brake applies and the same housing at the same time in a world where it does not apply. Only then would it be possible to accurately determine how the measure unfolds. However, this is simply not the case and it is always unclear whether a decrease in new rentals under new contracts can actually be attributed to the rent brake or whether other circumstances - unrelated to the rent brake - are responsible for the price decrease.

In empirical social research, however, there are various methods for quantifying the effects of such a reform nonetheless. Studies based on these methods come to the unanimous conclusion - in contrast to the price development shown above - that the reform had a price-dampening effect. Although this price dampening effect is relatively small.2

The evaluation is also made more difficult by a derogation. This is because even after the introduction of the rent brake, the new contractual rent of formally braked flats may always be raised up to the amount of the rent paid by the previous tenant. However, there are no meaningful data sets on the level of previous rents in Germany, which is why the legal rent and thus the potential effects of the reform always remain unknown. Moreover, even price damping effects reveal little about whether the intended target group of middle and low income households actually benefits from it.

In short, one knows surprisingly little.

1. The allocation was analogous to the definition of Thomschke (2019).
2. See for example the readable study by DIW (2018) oder Thomschke (2019a). [both in German]