Housing Standards 2007/2008:
The Factors behind the High Prices of Owner-Occupied Housing in Prague

4.1 Demand factors

4.1.2 How do young people behave in the residential real-estate market?

To test and better understand the strong preference for housing ownership among young people in Prague, an additional experiment, unique within the field of Czech housing research, was carried out (using software applications operating through a web interface). It is worth mentioning that this experiment has probably not yet been employed in the field of housing research, or at least we know of no other similar experiment conducted in the field of housing research in the world.

The experiment we conducted focused on testing the rationality of decisions made in searching for housing and the effect of additional information on how the residential market functions. The participants in the experiment were students and young people in general, usually still without their own permanent housing, who were recruited using calls for participants posted by university entrances. A total of 103 young people took part in the experiment. The experiment itself was performed using a software application specially designed for this purpose, simulating the development of the housing market in 28 half-year phases; the development of prices on the housing market was largely copied from the trend in the prices of residential real estate and rents in Finland during the last important price cycle at the start of the 1990s. During that time in Finland, a relatively significant increase in prices was followed by a sharp fall, which was so substantial that since then it has been one of the few countries in which the share of rental housing in the total housing stock has grown, and has done so while causing the share of owner-occupied housing to decrease. For the purpose of the experiment in the CR the development of prices and rent were differentiated into three different market segments in three Prague locations: flats in a newly constructed building, flats in an older brick building, and flats in a pre-fabricated panel building.

Each participant during each of the 28 phases (one phase equalling one half-year period) had a given level income, savings, essential expenditures on other consumption (than housing) of the household, and interest rate, could decide whether he/she would live in a rental flat or in owner-occupied housing; in either case he/she could decide between the three above-mentioned types of flat (locations) and four flat sizes (owing to the complexity of including it a family house was not one of the options). If they decided to buy a flat they could use their accumulated savings to do so (all the participants entered the experiment with a certain initial amount of savings and then increased their savings by the amount not spent on housing and other household consumption – and interest was earned on that sum), or with a mortgage loan, and in that case he/she was able to use a ‘mortgage calendar’ to calculate their monthly annuity mortgage repayments. At the end of the fixed interest rate term (after two phases, i.e. one year) they could always pay a part of or the entire mortgage with accumulated savings. If the participant opted for the mortgage loan they were in each phase informed of the amount of the annuity payment and the amount of the mortgage still payable.

At the very start each participant in the experiment could determine whether, when looking for their first permanent housing, they would be looking for housing alone or would be looking for housing with their partner; at the same time they could choose whether upon completing school and starting work they would be working in the commercial sector or the state sector. Based on these decisions the participants either entered the experiment as individuals or as a couple and as a commercial-sector employee or a state-sector employee. Given that the reactions to the simulated situation were not intended to be aimed at obtaining the highest possible gain from investments but rather were supposed to approximate as closely as possible real decision-making, participants in the experiment were rewarded equally regardless of what their final property was at the end of all 28 simulated phases.

Around one-half of the participants in the experiment were also given basic information on how the housing market functions, its cyclical trends, and the trend in the prices of residential real estate in a time series for selected advanced countries. One of the main objectives of the experiment was to trace whether the reactions of informed participants differed significantly from the reactions of participants who did not possess that information.

The results of the experiment confirmed findings thus far that there is a very strong preference for housing ownership among young people in the Czech Republic, and even additional information about the hidden risks involved in investing into housing ownership or the experience of a very deep and long-term drop in flat prices during market simulations does anything to alter this. ‘Informed’ participants in the experiment used their knowledge of the possibility of a deep and long-term drop in prices only to hesitate longer than the ‘uninformed’ participants in buying a flat, waiting to the point where prices bottomed out, and then they still bought flats. Both groups of participants (informed and uninformed) sooner or later decided to buy their housing, but informed participants were more cautious about making this decision. At the very start, when the simulated prices of flats rose they postponed buying their housing and often they only bought a flat once there was an incommensurate increase in prices (which led them to notable ‘losses’); on the other hand, if the prices fell they waited to buy a flat until the prices reached a low and bought their flats at the lowest prices (with the knowledge that the price cycles can be very deep; this conversely contributed to their ‘gains’). Although it was not the intention of the experiment, if we were to evaluate the ‘gain’ for the entire period of the experiment (the amount of accumulated wealth in the form of savings and the market value of the flat at the end of the experiment, after deducting the unpaid portion of the mortgage), then we would find no significant difference between the informed and the uninformed participants.

The informed participants in the experiment managed somewhat better during the crisis period of the price drop (or in the period when the prices were at their lowest). All those participants that displayed the ‘best’ strategy (leading to the greatest accumulated wealth) were from the group of ‘informed’ participants, while only one of those with the ‘worst’ strategy was from the group of informed participants. Also interesting is that purely from the perspective of the amount of accumulated wealth the best strategies proved to be ones that responded to the current situation in the market (even accepting the possibility of moving back into rental housing from owner-occupied housing), while conversely the strategy used most often in reality of gradually moving up the housing ladder into ownership turned out to be the least favourable in the case of a cyclical development in the housing market. The informed participants at least to start with tend to buy ‘better quality’ housing in a new building and avoid the ‘risk’ of housing in a pre-fabricated panel building. In the experiment, the simulated price drop was biggest in the case of panel flats, and so investment into panel flats hypothetically paid off the least. Uninformed participants in the end gained the same experience from the behaviour of the market itself and took this into account in subsequent steps they made; in some cases they even ‘outdid’ the informed participants in their subsequent actions.

The rationality of the decisions about which form of housing tenure to choose in individual phases of the experiment was also evaluated. Participants were regarded as acting rationally if in the given phase of the simulation they chose owner-occupied housing (flat ownership) when the annual user costs of owner-occupied housing were lower than the annual rent for a flat of the same size and the same type of construction; or if they chose rental housing when the amount of the annual rent was lower than the annual user costs for a privately owned flat of the same size and the same type of construction as the chosen rental flat. The user costs of owner-occupied housing were determined as the sum of the annual interest costs in buying housing (after taking into account the possibility of deducting interest from a mortgage loan from the income tax base according to rates in effect in 2007), the annual contribution to the building maintenance fund and the assumed costs derived from the depreciation of the flat reduced by the average annual nominal price appreciation of the flat that the owner would get assuming he/she held onto the given type of flat from the moment of purchase up to the the last phase of the experiment (28^th phase). In the situation where the price of the flat at the time it was bought was higher than the price of the flat in the last (28^th) phase (the respondent would end up with a capital loss), the user costs of owner-occupied housing were conversely increased by the amount of this loss.

The informed participants in the experiment did not generally act more rationally. However, although possessing information about how the market functions did not itself lead to more rational action on the part of those participants in the experiment generally (in all phases of the price cycle), the behaviour of the informed participants was significantly more rational than that of other participants during a period of crisis (a price crash).

One of the things that the results of the experiment showed is that young people in the Czech Republic are willing to pay a significant price to obtain owner-occupied housing: for couples with incomes from employment in the commercial sector, at the start of the experiment (in the first phase) the housing expenditure-to-income ratio in the rental sector was around 25%; nevertheless a large portion of those households were willing from the very start of their housing career to pay on average up to 42% of their total income in order to immediately obtain privately owned housing. Analogically, throughout the entire period of the experiment, households of couples working in the commercial sector paid on average only around 23% of their total income on rent and yet these households were willing on average to sacrifice, according to the weighted average, around 38% of the their net monthly income in order to buy their own flat.