From Jonas Schweizer and Gerd Kommer
The “new German residential real estate boom” began in 2010, which continues to this day and has led to stratospheric real estate valuations in some medium-sized and large cities in Germany. This boom was new a decade ago in that it followed a long “slow-motion crash” in residential real estate from 1981 to 2009. At the end of 2009, inflation-adjusted prices in Germany were 30% below those in 1980. Hardly anyone remembers that today.
The demand for residential real estate is enormous today after eleven years of price increases. It is fueled by the widespread overestimation of Return of the residential real estate asset class, from the false belief that returns in the recent past are a good return indicator for the future, from FOMO emotions (Fear of Missing Out), from the “theory” Real estate is a particularly safe investment, from the ambivalent assumption that real estate provides good protection against inflation, from the alleged lack of attractive investment alternatives and, of course, from historically uniquely low loan interest rates. The only thing that is curbing demand is the extremely high valuations and the hesitancy of many banks to provide financing with a debt share of 80% to 90%.
Be that as it may, most real estate investments by private households should begin with a cost-effectiveness calculation. At a minimum, the question that needs to be answered is whether purchasing a property – with or without external financing – is economically viable for the household. Such a calculation can include input variables such as the expected increase in value, the achievable rent (the actual rent for a rental property or the rent saved for an owner-occupied property), the risk of loss of rent, the debt service, the additional costs of the purchase and the maintenance costs, which are the subject of this blog post.
In our experience, sellers and buyers photoshop two input variables particularly often in such calculations: the forecast increases in value and the expected maintenance costs. The reason: Sellers and brokers have a natural interest in financially improving a proposed sales transaction with high assumed increases in value and/or low maintenance costs. Buyers photoshop when making such calculations because many of them are subject to an inner compulsion to consider their emotional dream object to be economically attractive to themselves and others.
Achieving agreement on what are appropriate assumptions for the future increase in value of a particular residential property is difficult because there is a lot of scope for completely contradictory views on market developments and demand in the long-term future and especially because everyone interprets market history according to their own taste. The inflation-adjusted increase in the value of German residential properties in the five years from 2016 to 2020 was a spectacular 7.4% p.a., and in the 45 years from 1971 to 2015 it was a paltry minus 0.2% p.a.
However, when it comes to maintenance costs - unlike price increases - a little sober research leads to easily verifiable guideline values, where there is relatively little room for subjective discussion. However, because these benchmarks derived from factual logic and history are significantly higher (less attractive) than what the parties involved want based on their interests mentioned above, lower numbers are still often used. We will examine the background below.
In real estate advice books, in real estate blogs and by building finance brokers, the benchmark figure for maintenance costs for residential properties has been quoted since Anno Domini, between six and twelve euros per square meter of living space per year. It is impossible to understand where these numbers come from and how they are justified. They were probably postulated at some point by a nameless real estate expert and then continued without verification by each new generation of “experts”. Why these numbers should be static, i.e. not increase over time with construction cost inflation, is also a mystery.
In Cologne, the square meter of living space cost an average of 4,200 euros in 2020. This made the Rhine metropolis the cheapest of the seven cities in Germany with over 500,000 inhabitants. Assuming that 85% of these costs are attributable to the building, six euros per year results in a ridiculously low maintenance cost ratio of 0.17% per year. Even twelve euros would only be 0.34%.
But the Photoshopping of maintenance cost estimates doesn't stop there. There are recommendations circulating in the real estate industry's marketing publications regarding the level of maintenance cost reserves for apartments in apartment buildings, which are even lower than the six to twelve euros mentioned above. Then the regular lack of information no longer plays a role that the reserve to be paid to the homeowners' association only relates to the common property (outer walls, roof, stairwell, etc.) - excluding special property (the larger part in terms of value).
Absolute maintenance cost guidelines in “so many euros per square meter” are absurd because they take neither the age-related condition nor the value of the property into account. For example, you can imagine two new, different 180 sqm single-family homes in the same micro-location. The construction costs of one were 400,000 euros, those of the other 800,000 euros. The first property was built simply and cheaply, the second was built to a high standard and expensively. It is obvious that two houses of the same size but very different in quality will not have the same maintenance costs in the future, assuming that the owner wants to compensate for the same percentage loss in physical value per year for both apartments.
A somewhat more realistic benchmark for maintenance costs is the number “1% per annum” sometimes mentioned in the advice literature, based on the proportion of the property value that is attributable to the building. However, it is often left vague that the building value here means the gradually increasing current value, not the static value when purchased. Regardless, this 1% is very likely still too low. Why can be made plausible as follows.
(1) Let's start with common sense: At 1% p.a., the building (assuming full compensation for the loss in value through maintenance) would have to be habitable for 100 years without this maintenance (1% linear loss in value per year = 100 years until the value has fallen to zero). The absurdity of this assumption in relation to the quality of the average residential building today, not to mention people's changing floor plan preferences over the decades, needs no further proof.
(2) In Germany, the “deduction for wear and tear” (Depreciation) of 2.0% p.a. and the actual annual maintenance costs are tax deductible for rental properties. The sum of these two positions is over 3.0% in the long term. This is no coincidence. Rather, it is the political result of negotiations between property owners, who for obvious reasons want the deduction rates to be as high as possible, and the state, which wants the lowest possible values for equally obvious reasons. You meet where the arguments of both sides are approximately equally good, i.e. h. close to the reality of actual annual loss of value from physical wear and aging.
(3) According to the “Peters formula” known in the real estate industry, the estimated annual maintenance costs in a normal apartment building are 1.88%, based on the initial value of the building part (manufacture costs). Detached houses (single-family houses, semi-detached houses, terraced houses) are, on average, subject to a higher physical loss of value due to technical and construction reasons (see the keyword “Peters formula” in the German Wikipedia).
(4) Vonovia SE, Düsseldorf, Germany's largest apartment owner and DAX member with almost 500,000 units at the end of 2019, had an average maintenance cost ratio of 1.2% p.a. for the five financial years 2015 to 2019. If the calculation of the rate is not based on the “normal market value” of the housing stock stated in the annual report, but rather the significantly lower “adjusted market value” mentioned there, the average maintenance rate increases to 2.4% p.a. The truth is probably somewhere between 1.2% and 2.4%. It can be assumed that, due to its size, Vonovia can carry out maintenance at least a third more cost-effectively than a single homeowner. This is offset by an increased loss of wear and tear for rented apartments compared to owner-occupied homes, but this is likely to be largely offset by the fact that the Vonovia apartments are structurally much simpler than the average home in Germany. (The simpler an apartment is, the lower the percentage maintenance costs tend to be.) In addition, apartments in apartment buildings (the largest part of Vonovia's portfolio) cause lower maintenance costs than detached single-family homes, semi-detached houses or terraced houses. If all of the adjustment factors mentioned were quantified relative to the situation of small landlords or owner-occupiers, the range of 1.2% to 2.4% would tend to go up.
(5) In an essay by two economists from the Federal Reserve Bank of Cleveland, a regional American central bank which - unlike banks, building societies and building finance brokers - is not subject to a conflict of interest, an annual maintenance rate of 2.0% based on the current property value including land is suggested as a "good benchmark" (Ergungor/Zaman 2011). Since the average building quality as well as the construction costs in the USA are probably lower than in Germany, one can assume a rate for Germany - based on the part of the building - that is approximately a third lower (with a part of the building of 85%, this would be 1.57% (= 2% ÷ 85% × ⅔). Two academic US real estate economists (Aked/Marutscho 2016) arrived at approximately the same figure in an essay for rented properties (Offices, retail, restaurants, living) in the USA. The British real estate economists Chambers and others also made a similar assessment in 2020 for residential real estate in Great Britain.
If you use a cost ratio of 1.5% p.a. on the current value of the property as the average estimate for a normal apartment and assume that the building accounts for 90% of the total costs, then the owner must expect an average of 1.5% × 90% ≈ 1.4% maintenance costs per year. Over a period of 30 years, around 40% of the current value will be spent on repairs.
This cost ratio includes everything that goes into repairs and maintenance - in the case of an apartment, any maintenance reserves, but not expenses for an administrator. Expenses for modernization or expansion are not maintenance because they create something “new” or “improved”.
Of course, it is possible to only spend an average of half a percent per year or even less on maintenance for 20 years, for example. However, the inevitable consequence is that the physical quality, and thus the residential use of the property, will noticeably decline over these 20 years. If this is the case, the estimated value for the development of the property must be corrected downwards accordingly in a future-oriented economic analysis. The estimated values for property price developments and the amount of additional costs are directly and closely related to each other. If maintenance is set too low, the price increase estimate must also be reduced accordingly.
For a specific residential property, what characteristics influence the maintenance cost ratio or the maintenance costs per square meter of living space upwards or downwards within the general, medium ranges mentioned in this section?
- Detached homes have maintenance cost ratios 10% to 30% higher than apartments.
- Very old buildings, all other things being equal, have a maintenance cost ratio that is around 10% to 20% higher than younger buildings.
- Buildings in poor structural condition – as long as this condition is not remedied through comprehensive renovation – have a 10% to 30% higher maintenance cost ratio than buildings in good structural condition without a maintenance backlog.
- High-quality, expensive buildings have a slightly higher maintenance cost percentage than simple, inexpensive buildings.
- Buildings in areas with high humidity, e.g. B. near a seashore, have higher maintenance cost ratios than buildings in areas with low humidity.
- Buildings in large cities have higher maintenance cost ratios than buildings in rural areas because labor costs for tradesmen in cities are higher.
- Listed buildings have significantly higher maintenance cost ratios than non-listed buildings.
In the long-term future, an increase in maintenance costs that goes beyond the general inflation of manufactured goods is to be expected because the German state tightens the building regulations in the areas of energy, environmental protection, health protection and accessibility of living spaces for people with disabilities almost every year.
Conclusion
The additional cost guideline of six to twelve euros per square meter per year or 1% per year, which representatives of the real estate industry often spread to owner-occupiers or small landlords, may be good real estate marketing, but hardly realistic estimates for a future-oriented profitability calculation or return calculation for a typical medium-quality apartment. For this we should set a rate of not less than 1.5% of the fair value of the part of the building. For a property in poor construction condition, for very high-quality properties and for detached houses, a maintenance cost rate of 1.7% to 2.5% p.a. for the part of the building is generally more appropriate.
literature
Aked, Michael/Marutscho, Jim (2016): “Next Season’s Meager Harvest in Commercial Real Estate”; Internet reference: here
Chambers, David/Spaenjers, Christophe/Steiner, Eva (2020): “The Rate of Return on Real Estate: Long-Run Micro-Level Evidence”; Internet reference: here
Ergungor, Emre/Zaman, Saeed (2011): “Buy a Home or Rent: A Better Way to Choose”; Internet reference: here