← Összes podcast
2023/1. lapszám · Makrogazdaság és gazdaságpolitika

Factors influencing the volume of residential property transactions in Romania during the Covid-19 pandemic

Oborocea Mónika

Megjelent: 2023 · Közgazdász Fórum, 2023/1. lapszám · Oldalak: 35–57

„Mi történt a romániai lakáspiaccal a járvány alatt?”

Meghallgatom a podcastot

Bővebb beszélgetés · 9 perc
HU EN A podcast nyelve: magyar · EN: az angol változat még nem készült el
0:00-8:56

Megjelent: 2023 · Közgazdász Fórum, 2023/1. lapszám · Oldalak: 35–57

A teljes cikk HU EN

OJS ↗ A teljes lapszám (PDF) ↗

This English text is a machine translation of the original Hungarian article, made with a local AI model. The original is authoritative: read the original (Hungarian)

Közgazdász Fórum · Issue 1/2023 · 35–57

Factors influencing the volume of residential property transactions in Romania during the Covid-19 pandemic

Abstract. In Romania, the volume of residential property transactions increased alongside the rise in new Covid-19 infection numbers during the pandemic. This study examines the underlying drivers of this increase during the period between April 2020 and October 2022. The results of the panel regression indicate that the unemployment rate had the strongest impact on changes in transaction volume, with a negative correlation. The number of sales was also influenced by the interest rate on residential loans denominated in RON and the construction cost index, both of which exhibited negative correlations. The findings demonstrate that during the pandemic, transaction volume followed the movement of the economic sentiment index through a direct relationship between the two variables; thus, the sentiment index may serve as an effective predictor of trading volume. No significant relationship was found between the volume of residential property transactions and net real wages. One possible reason for this is that, due to government measures (such as technical unemployment), the average net real wage in Romania did not decrease significantly during the period under investigation.

Keywords: Covid-19, real estate market, residential property transaction volume

Introduction

In contrast to previous economic crises, an unusual phenomenon occurred in

Romania during the Covid-19 pandemic, where the transaction volume of residential properties increased significantly between 2020 and 2022, despite deteriorating economic indicators. Concurrent with a rising unemployment rate and declining GDP, the transaction volume of residential properties in 2020 was 8 per cent higher on an annual basis compared to 2019, even though the National Agency for Cadastre and Real Estate Advertising (ANCPI, Agenția Națională de Cadastru și Publicitate Imobiliară) recorded a 28 per cent decrease in the second quarter of 2020 compared to the same quarter in 2019. This means that residential property transactions in the fourth quarter of 2020 were nearly 40 per cent higher than in the corresponding quarter of the previous year. This increasing trend continued through 2021 and 2022.

Furthermore, according to data from the National Bank of Romania (BNR), a significant portion of these increasing transactions was financed by household credit. The volume of residential mortgages disbursed in lei for housing purchases was 0.9 per cent higher in 2020 compared to 2019, and by 2021, it was already 43 per cent higher than in 2020.

The upward trend did not halt in 2022; households took out 13 per cent more residential mortgages compared to the previous year, even as the Romanian consumer price index approached 16 per cent on an annual basis by the end of 2022. According to the central bank's report (BNR 2022), in the period between January and September 2022, households financed 38 per cent of residential property transactions through credit.

The present study examines the drivers behind this growth in residential property transactions occurring within such an economically unfavourable environment. The research is unique in that no study has hitherto been conducted that employs econometric models to examine the evolution of residential property transaction volumes in Romania during the Covid-19 pandemic. The current research is based on econometric estimations and tests performed in the R statistical programming language within the RStudio integrated development environment. The research topic may be of outstanding importance not only for real estate agencies but also for the general population, as 95 per cent of Romanians live in their own homes, according to the latest 2022 data from Eurostat (Eurostat 2023). We are at the forefront of this in the European Union, where the EU average is 69 per cent.

Following the literature review, the data and their descriptive statistical characteristics are presented, followed by the methodology. After the analysis of the results, the study concludes with the derivation of conclusions.

Literature Review

The literature examines the volume of residential property transactions during the Covid-19 pandemic according to the following key variables: new infection numbers, the inflation rate (construction cost index), interest rates, the economic sentiment index, the unemployment rate, and income.

The impact of the Covid-19 pandemic on the volume of residential transactions has been a subject of debate in the literature. On one hand, rising Covid cases, the fear of infection, and social distancing regulations restrained household consumption (Pichler et al. 2020; Baker et al. 2020), and consequently, residential property transactions; on the other hand, pandemic-related lockdowns and the rate of disease spread in densely populated settlements (supplemented by fiscal and monetary incentives) overrode the negative pressure on housing demand and supply (Diamond et al. 2022; Gallent et al. 2023).

Another characteristic of the relationship between transaction volume and the number of cases is that real estate sales decreased without exception when official lockdowns were introduced to curb the pandemic, but began to increase following the lockdowns (Liu–Tang 2021; Tsai et al. 2022). In several countries, this growth was so significant that it was able to compensate for the decline in demand during the lockdowns on an annual basis. This phenomenon also occurred in the US real estate market (Gamber et al. 2023).

The impact of inflation on transaction volume is also a subject of debate. Examining the Swedish real estate market, Malmström and Schultz (2017) demonstrated that rising inflation signifies increasing transaction volume. The authors investigated the economic factors influencing foreign direct investment in the Swedish real estate market using the 2016 transaction data of a Swedish real estate agency, Newsec.

Based on their results, in a high-inflation environment, direct investments in the real estate market serve as a hedge against the depreciation of the purchasing power of money.

The inverse relationship between inflation and transaction volume operates through several channels. Follain (1982) established that a high inflation rate increases nominal interest rates, and consequently, the nominal instalments of mortgage loans. According to Dekimpe and van Heerde (2023), this leads to fewer households being able to afford borrowing, thereby reducing the number of housing transactions. This phenomenon occurred in the European real estate market during the Covid-19 pandemic, when high inflation in 2021–22 forced central banks to increase policy rates (ECB 2023). As a result, mortgage interest rates doubled in most European countries in 2022, while in Finland, Slovakia, Switzerland, and the United Kingdom, they tripled.

Inflation can also reduce demand for real estate through wages.

High inflation reduces the real disposable income of households, and declining income is directly related to transaction volume (Ortalo-Magné–Rady 2004). Valderrama et al. (2023) pointed out that in the case of variable-rate mortgage loans, the decline in households' real income and the rise in mortgage interest rates increase the default risk of borrowers; the latter negatively affects bank capital, which consequently reduces the volume of loans granted.

Using American and British datasets, Meen (2002) demonstrated that the construction cost index, which follows the trend of inflation, is inversely related to house prices, and rising prices lead to decreasing demand (Stein 1995; Andrew–Meen 2003; Akkoyun et al. 2013).

The impact of the unemployment rate on residential transaction volumes is inconclusive. Gan et al. (2018) found that an increase in the unemployment rate led to higher house prices and fewer sales transactions in the Texas real estate market. The authors pointed out that a rising unemployment rate influences the size of the real estate market on both the demand and supply sides: on the demand side, it serves as a financial constraint and reduces the number of buyers; on the supply side, due to the rising unemployment rate, sellers wait to transact, knowing that demand is decreasing. Clayton et al. (2008) also confirmed an inverse relationship between the unemployment rate and transaction volume, and explained the direct relationship between house prices and the unemployment rate as follows: as a result of the rising unemployment rate, sellers themselves face financial difficulties, thereby increasing property prices.

Contrary to the above, Slavata (2021) pointed out that with increasing unemployment, the number of dwellings offered for sale or rent increases, as the intention to move among jobseekers rises due to a tight labour supply.

According to the author, changes occurring in the labour market have a real-time effect on the housing market.

Several studies have addressed the economic sentiment index as a proxy variable for GDP trends (Guzmán 2009). Kitrar's (2021) study, based on Russian data from 1998–2020, demonstrated that gross domestic product (GDP) and the economic sentiment index move closely in tandem.

The direct relationship between the economic sentiment index and sales volume is undisputed. In a study by Białowolski (2019) examining the relationship between the economic sentiment index and the financial decisions of Polish households, it was found that consumer sentiment influences household borrowing and saving behaviour. If consumer confidence increases, so does the belief that the economy is performing well, which in turn increases the willingness of households to take out credit for durable consumer goods, including house purchases. This conclusion is consistent with the finding of Lamdin (2008) that rolling credit generally increases alongside rising consumer sentiment. Hui and Wang (2014) demonstrated that since real estate demand is driven in part by the sentiment of market participants, the market sentiment index can be an effective predictor of both price and trading volume.

Very few studies processing Romanian data have emerged that examine residential sales volumes using econometric models.

Enăchescu and Bănică (2019) used linear regression to estimate the trend of real estate sales in Romania between 2009–2018. In the authors' estimation, the explanatory variables were the permanent resident population, income, and GDP per capita. Although the authors were unable to demonstrate a linear relationship between the variables, they succeeded in identifying growth and decline trends through the descriptive statistical characteristics of the transaction volume.

Ion et al. (2021) also utilised descriptive statistical characteristics to examine the impact of the Covid-19 pandemic on the construction industry and real estate transactions. Using quantitative data, the research supports the decline in demand for new constructions and the decrease in the number of real estate transactions during the period of official lockdowns.

Apart from real estate transactions, several studies have analysed property prices. Moroșan et al. (2020) examined how the number of residential mortgages and the level of household income influenced housing prices between 2014–2019. According to the results, both variables influence housing prices, yet the study does not answer through which channels the effect of the explanatory variables is exerted. A summary of the literature review is provided in Table 1. Table 1: Summary table of the literature review. Investigated | Identified effects | Region | The impact of Covid cases on residential transactions | Rising Covid cases, fear of infection, and social distancing rules deterred households from consumption and residential transactions. | Pichler et al. 2020 UK | Baker et al. 2020 USA

Diamond

The pandemic-induced lockdowns and the rate of disease spread in densely populated settlements overrode the negative pressure on housing demand and supply in the 2022 USA et al. Gallent et al. 2023 UK.

Liu–Tang 2021 China | Real estate transactions decreased without exception when authorities introduced lockdowns to curb the pandemic, but began to increase after the lockdown. | Tsai et al. 2022 China | Gamber et al. 2023 USA

The impact of the pandemic on the construction industry and real estate transactions was examined using descriptive statistical characteristics. Ion et al. 2021 Romania The research supports with quantitative data the decline in demand for new constructions and the decrease in the number of real estate transactions during the period of official lockdowns.

Author

Table or figure – see the PDF version of the issue for its exact content.

The impact of the permanent resident population, income, and GDP per capita on the development of real estate sales was examined. Although the authors were unable to establish a correlation between the variables, they succeeded in identifying growth and decline trends through the descriptive statistical characteristics of the transaction volume.

It was investigated how the number of residential mortgages for property purchases and the magnitude of household income influenced house prices between 2014 and 2019. According to Romania et al. (2020), both variables influence house prices, yet the study does not provide an answer as to through which channels the effect of the explanatory variables is exerted.

The impact of the unemployment rate on residential transaction volumes (Gan et al., 2018, USA): An increase in the unemployment rate leads to higher house prices and fewer sales transactions (Clayton et al., 2008, USA).

Table or figure – see the PDF version of the issue for its exact content.

Identified effects

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

As no study using Romanian data has yet been conducted that estimates the impact of the Covid-19 pandemic on residential transaction volumes using econometric models, the present study seeks to address this gap. Based on the literature, the hypotheses formulated in this study are as follows:

H1: Covid-19 case numbers have a positive effect on the residential transaction volume in Romania.

H2: The inflation rate has a negative effect on the residential transaction volume in Romania.

H3: Mortgage interest rates have a negative effect on the residential transaction volume in Romania.

H4: The unemployment rate has a negative effect on the residential transaction volume in Romania.

H5: Household income has a positive effect on the residential transaction volume in Romania.

H6: The economic sentiment index has a positive effect on the residential transaction volume in Romania.

Data and methodology

Data and descriptive statistics

In the estimated model, the dependent variable is the monthly number of residential transactions published by the ANCPI. The explanatory variable is the number of new infection cases, while the control variables are the national average net real wage, the construction cost index, the interest rate on variable-rate mortgage loans, the harmonised unemployment rate, and the economic sentiment index (the GDP proxy variable).

The period under investigation is the period between April 2020 and October 2022. The selection of this period is justified by the availability of data: the National Institute of Public Health published monthly continuous county-level data regarding new infection cases between April 2020 and October 2022.

In order to ensure that the county-level data were comparable, a transformation of the transaction volume and Covid-19 case number variables was required. Accordingly, monthly county-level data on residential transaction volumes were reasonably projected per 1,000 inhabitants based on formula 1. The population figures for the counties were provided by the 2021 census data from the National Institute of Statistics. T_TRANS_UNIT =

Table or figure – see the PDF version of the issue for its exact content.

I proceeded in the same manner with the new infection case count variable; in contrast to Figure 1, I replaced the monthly transaction volume with the monthly number of new infection cases in the given county. The data shown in Figures 1 and 2 present the county averages projected onto the period under investigation.

The figures prepared for the remaining variables can be viewed in the appendix. Table 2 provides a summary of the variables examined.

A summary of the descriptive statistical characteristics of the variables can be found in Table 3, with the number of observations being 1302. In the case of transaction volume and Covid case numbers, there is a significant discrepancy between the median and the mean, and the standard deviation is large. Since the values for these two variables do not follow a normal distribution, "within" estimation is required for panel regression models (Bramati–Croux 2007). Based on the median and mean values, the income, unemployment rate, interest rates, construction cost index, and economic sentiment index variables follow a normal distribution.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Methodology

Since both cross-sectional and time-series observations are present, the literature primarily recommends panel regression models for the rejection or retention of hypotheses (Phillips–Han 2019; Yilmazkuday 2023). The steps for estimating panel regression models (Baltagi 1998) are: · testing the stationarity of the variables (Dickey–Fuller 1981),

· estimating the models (pooled ordinary least squares, fixed effects, and random effects models), · model selection using the F-test, the Breusch–Pagan Lagrange Multiplier test, and the Hausman (1978) test, · model diagnostic tests (cross-sectional dependence, autocorrelation, and heteroscedasticity).

The Pooled Ordinary Least Squares (POLS) model is the simplest panel regression model, but it does not account for heterogeneity within the group or within the time dimension (Pedroni 2001). Its general formula can be written as follows:

Yit = $\alpha$ + $\beta$1Xit + $\varepsilon$it,

Table or figure – see the PDF version of the issue for its exact content.

where $\alpha$ is a constant, $Y_{it}$ is the dependent variable, $t$ is the time dimension, $i$ is the cross-sectional dimension, $X_{it}$ is the independent variable, $\beta_1$ is the coefficient of the independent variable, $\varepsilon_{it}$ is the error term for the full regression, $i = 1, \dots, N$.

The Fixed Effects (FE) model is based on the assumption that every observed unit possesses its own unique characteristics. We assume that these characteristics have an effect on the explanatory or dependent variable. This assumption underlies the correlation that exists in the FE model between the explanatory variable and the unobservable, group-specific error term. If we remove this time-invariant characteristic from the explanatory variable, we obtain the net effect of the explanatory variable (Stock–Watson 2004; Bartels 2008). If we wish to account for time-invariant fixed effects, the model can be expressed as follows: $Y_{it} = \alpha + \beta_1X_{it} + u_i + \varepsilon_{it}$,

Table or figure – see the PDF version of the issue for its exact content.

where $u_i$ is an unobservable, group-specific error term, $i$ being the group or cross-sectional dimension, $i = 1, \dots, N$.

If there is no correlation between $u_i$ and $X_i$, then a random effects model can be applied. The random effects model may be preferable if the differences between groups are random (Bell–Jones 2015). The general formula for a model estimating individual-specific, but time-invariant, random effects is:

$Y_{it} = \alpha + \beta_1 X_{it} + (u_i + \varepsilon_{it})$,

Table or figure – see the PDF version of the issue for its exact content.

Model selection tests include the Hausman (1978) test and the F-test (Baltagi 1998). The Hausman test assists in choosing between fixed-effects and random-effects models, with the following hypothesis:

Table or figure – see the PDF version of the issue for its exact content.

According to the global F-test results of the estimated POLS, fixed-effects, and random-effects models (see Table 6), all three models are valid. The adjusted $R^2$ indicates the percentage of the variance in residential property transaction volume explained collectively by the independent variables. Since the POLS model is valid, the Variance Inflation Factor (VIF) must be calculated to analyse multicollinearity between the explanatory and control variables, the results of which are presented in the following table. Based on the VIF values, all variables could remain in the estimation.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

Based on the adjusted $R^2$, POLS is the best of the three models, yet model selection tests indicated otherwise. According to the results of the F-test and the Hausman test, the fixed effects model is the most appropriate for estimation. Since the observations are highly heterogeneous for two variables in the database (the number of new infection cases per thousand people, and the volume of housing transactions per thousand people) and there are outliers that may cause significant deviations from the true parameters, a within estimator must be applied to ensure the robustness of the estimated parameters (Bramati–Croux 2007). The t-tests performed on the model estimated in this manner, and their results, are presented in Table 8. Table 8: Results of the t-tests for the fixed effects model Variables

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

According to the results of the model diagnostic tests, there is cross-sectional dependence between groups in the fixed effects model, and the error terms are autocorrelated and heteroskedastic. Since the results of the estimated fixed effects model are biased according to the model diagnostic tests, it was necessary to perform an Arellano transformation (Arellano–Bond 1991) and re-estimate the model. The results of the t-tests following the transformation are presented in Table 9. Table 9: Fixed effects model results after Arellano transformation Variables

Estimated coefficients

Table or figure – see the PDF version of the issue for its exact content.

Table or figure – see the PDF version of the issue for its exact content.

In the post-transformation model, net average wage was not significant, whereas the loan interest rate was significant at 8.15 per cent; for the remaining variables, the significance level was below 5 per cent. Based on the signs of the estimated coefficients, it can be stated that Covid-19 case numbers and the economic sentiment index had a positive effect on the transaction volume of dwellings. The number of sales was negatively influenced by the unemployment rate, the loan interest rate, and the construction cost index.

The formula for the fixed effects model is:

T_TRANS_UNITti = 0,002 * TCOVID19 ti – 0,066 * UNEMPRATE – 0,009 * F2INTRATE – ti

Table or figure – see the PDF version of the issue for its exact content.

Conclusions

The results of the estimated fixed-effects model show that in Romania, between April 2020 and October 2022, an increase in the number of new infections, as well as the economic sentiment index, was accompanied by an increase in the volume of residential property transactions. Conversely, an upward movement in the interest rate of variable-rate mortgage loans, the construction cost index, or the harmonised unemployment rate caused a decrease in the number of housing transactions.

The results are consistent with the findings of Diamond et al. (2022) and Gallent et al. (2023), suggesting that the higher the number of new infections during the examined period, the more people purchased housing. The pandemic-induced lockdowns and the rate of disease spread in densely populated settlements (cities)—supplemented by fiscal and monetary incentives—overrode the negative pressure on housing demand and supply in Romania.

The number of residential property transactions in Romania was inversely related to the mortgage interest rate and the construction cost index, the latter of which acts as a proxy variable for inflation (Meen 2002; Akkoyun et al. 2013). In Romania, between April 2020 and October 2022, the increase in the inflation rate did not incentivise a significant portion of the population to invest in the real estate market, in contrast to, for example, the Swedish population (Malmström–Schultz 2017); rather, the rise in nominal interest rates meant that fewer households could afford to take out loans, thereby reducing the number of housing transactions, similar to the results of Dekimpe and van Heerde (2023). This conclusion may be of particular importance for, for instance, real estate agencies and companies specialising in property development, especially if the government and the central bank implement measures during an economic crisis in an attempt to compensate for rising mortgage interest rates.

Following the panel regression estimation, it can be stated that during the examined period, the number of residential property transactions in Romania was inversely related to the unemployment rate, in accordance with the findings of Gan et al. (2018), and among all variables, the unemployment rate had the greatest impact on the dependent variable. This effect should also be considered by natural persons wishing to buy or sell property during an economic crisis, as a sustained increase in the unemployment rate will likely lead to fewer people engaging in housing transactions. The growth in the volume of housing transactions was also influenced by the fact that the government kept the unemployment rate relatively low through subsidies provided to employees and employers via so-called technical unemployment (whereby employers did not dismiss their staff in exchange for state compensation, despite them performing little or no actual work). During the examined period, the Romanian harmonised unemployment rate was 5.83 per cent, with the indicator reaching its maximum of 6.90 per cent in a single month, June 2020. In contrast, according to Eurostat (2023) data, the average harmonised unemployment rate in the European Union remained persistently at or above 7.5 per cent from the second quarter of 2020 until the end of the first quarter of 2021, only approaching the Romanian average of 5.83 per cent at the end of the examined period, from the first quarter of 2022.

In Romania, during the pandemic, transaction volume followed the movement of the economic sentiment index; thus, the sentiment index can serve as an effective predictor of trading volume, in accordance with the findings of Białowolski (2019) and Hui and Wang (2014).

The results show that of the five hypotheses formulated in the study, five could be confirmed following the model estimation. One hypothesis (H5:

household income has a positive effect on the volume of residential property transactions in Romania) had to be rejected, as no significant relationship could be established between net income and the volume of residential property transactions. One possible reason for this is that, due to government measures, the average net wage in Romania did not decrease during the period under investigation.

The limitations of the research include the fact that it does not examine how Covid case numbers affected types of real estate other than residential properties. Furthermore, it does not examine the behaviour of the home renovation market during the pandemic, which is closely linked to residential property sales and construction. A further extension of the research could be an analysis of how the sale and purchase of plots of land evolved in Romania during the pandemic. It is probable that significantly more people in Romania purchased land in rural areas during the pandemic than in the years preceding or following it, due to mobility restrictions.

References

Akkoyun, H. C.–Arslan, Y.–Kanik, B. 2013. Housing prices and transaction volume.

Journal of Housing Economics 22(2), 119–134.

Andrew, M.–Meen, G. 2003. House Price Appreciation, Transactions and

Structural Change in the British Housing Market: A Macroeconomic Perspective. Real Estate Economics 31(1), 99–116.

Arellano, M.–Bond, S. 1991. Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations. The Review of Economic Studies 58(2), 277–297.

Baker, S. R.–Farrokhnia, R. A.–Meyer, S.–Pagel, M.–Yannelis, C. 2020. How Does Household Spending Respond to an Epidemic? Consumption during the 2020 COVID-19 Pandemic. The Review of Asset Pricing Studies 10(4), 834–862.

Baltagi, B. H. 1998. Panel Data Methods. In: Ullah, A.–Giles, D. E. A. (eds.) Handbook of Applied Economic Statistics. New York: Marcel Dekker Publishing, 291–323.

Bartels, B. L. 2008. Beyond „fixed versus random effects”: a framework for improving substantive and statistical analysis of panel, time-series cross-sectional, and multilevel data. The Society for Political Methodology 9, 1–43.

Bell, A.–Jones, K. 2015. Explaining Fixed Effects: Random Effects Modeling of Time-Series Cross-Sectional and Panel Data. Political Science Research and Methods 3(1), 133–153.

Białowolski, P. 2019. Economic sentiment as a driver for household financial behavior. Journal of Behavioral and Experimental Economics 80, 59–66.

Bramati, M. C.–Croux, C. 2007. Robust estimators for the fixed effects panel data model. The Econometrics Journal 10(3), 521–540.

Clayton, J.–Miller, N.–Peng, L. 2008. Price-volume Correlation in the Housing Market: Causality and Co-movements. The Journal of Real Estate Finance and Economics 40, 14–40.

Choi, I. 2001. Unit root tests for panel data. Journal of International Money and Finance 20, 249–272.

Dekimpe, M. G.–van Heerde, H. J. 2023. Retailing in times of soaring inflation: What we know, what we don’t know, and a research agenda. Journal of Retailing 99(3), 322–336.

Diamond, W.–Landvoigt, T.–Sanchez, G. 2022. Printing Away the Mortgages: Fiscal Inflation and the Post-Covid Housing Boom. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4283119, accessed: 13 March 2023.

Dickey, D. A.–Fuller, W. A. 1981. Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root. Econometrica 44, 1057–1072.

Enăchescu, P. C.–Bănică, C. 2019. Analysis Of The Real Estate Market In Romania From The Point Of View Of The Number Of Transactions During 2009-2018. Scientific Bulletin-Economic Sciences 18(3), 39–46.

European Central Bank 2023. Economic Bulletin Issue 5. https://www.ecb.europa.eu/press/economic-bulletin/html/eb202305.en.html#:~:text=Energy%20prices%20fell%20again%2C%20dropping,%2C%20from%205.8%25%20in%20May, accessed: 12 December 2023. Eurostat 2023. Housing in Europe – 2023 interactive publication. https://ec.europa.eu/eurostat/en/web/products-eurostat-news/w/wdn-20231130-1, accessed: 30 November 2023. Follain Jr, J. R. 1982. Does Inflation Affect Real Behavior: The Case of Housing. Southern Economic Journal 48, 570–582.

Gallent, N.–Stirling, P.–Hamiduddin, I. 2023. Pandemic mobility, second homes and housing market change in a rural amenity area during COVID-19 – The Brecon Beacons National Park, Wales. Progress in Planning 172, 100731.

Gamber, W.–Graham, J.–Yadav, A. 2023. Stuck at home: Housing demand during the COVID-19 pandemic. Journal of Housing Economics 59, 101908.

Gan, L.–Wang, P.–Zhang, Q. 2018. Market thickness and the impact of unemployment on housing market outcomes. Journal of Monetary Economics 98, 27–49.

Guzmán, G. 2009. Using sentiment surveys to predict GDP growth and stock returns. In: Klein L. R. (ed.) The Making of National Economic Forecasts. Massachusetts: Edward Elgar Publishing, 319–348.

Hausman, J. A. 1978. Specification Tests in Econometrics. Econometrica 46, 1251–1271.

Hui, E. C.–Wan, Z. 2014. Market sentiment in private housing market. Habitat International 44, 375–385.

Ion, A.–Cezar-Petre, S.–Costin, C.–Elena, I.–Ovidiu, D. 2021. The Impact of COVID-19 on the Building Industry and on Real Estate Transactions in Romania. In: Dima, A. M–Anghel, I.–Dobrea R. C. (eds.) Economic Recovery After COVID-19. Bucharest: 3rd International Conference on Economics and Social Sciences, ICESS 2020, 157–174.

Kitrar, L. 2021. The relationship of economic sentiment and GDP growth in Russia in light of the Covid-19 crisis. Entrepreneurial Business and Economics Review 9(1), 7–29. Lamdin, D. J. 2008. Does Consumer Sentiment Foretell Revolving Credit Use? Journal of Family and Economic Issues 29, 279–288.

Levin, A.–Lin, C. F.–Chu, C. S. J. 2002. Unit root tests in panel data: asymptotic and finite-sample properties. Journal of Econometrics 108(1), 1–24.

Li, G.–Pengfei, W.–Qinghua, Z. 2018. Market thickness and the impact of unemployment on housing market outcomes. Journal of Monetary Economics 2018(98), 27–49.

Liu, Y.–Tang, Y. 2021. Epidemic shocks and housing price responses: Evidence from China’s urban residential communities. Regional Science and Urban Economics 89, 103695.

Maddala, G. S.–Wu, S. 1999. A Comparative Study of Unit Root Tests with Panel Data and a New Simple Test. Oxford Bulletin of Economics and Statistics 61(S1), 631–652. Malmström, D.–Schultz, A. 2017. Foreign Direct Investments on the Swedish Real Estate Market: Fundamentals Impact on Foreign Transaction Volume. https://www. diva-portal.org/smash/record.jsf?pid=diva2%3A1127714&dswid=3618, accessed: 05 Dec 2023. Meen, G. 2002. The Time-Series Behavior of House Prices: A Transatlantic Divide? Journal of Housing Economics 11, 1–23.

Moroșan, G.–Condratov, I.–Filipciuc, L. 2020. Influencing factors on housing prices in Romania. The USV Annals of Economics and Public Administration 20, 115–128. Ortalo-Magné, F.–Rady, S. 2004. Housing transactions and macroeconomic fluctuations: a case study of England and Wales. Journal of Housing Economics 13, 287–303. Pedroni, P. 2001. Fully modified OLS for heterogeneous cointegrated panels. In: Baltagi B. H.–Fomby T. B.–Hill R. C. Nonstationary panels, panel cointegration, and dynamic panels. Emerald Group Publishing, 93–130.

Phillips, P. C. B.–Han, C. 2019. Dynamic panel GMM using R. Handbook of Statistics 41, 119–144.

Pichler, A.–Pangallo, M.–del Rio-Chanona, R. M.–Lafond, F.–Farmer, J. D. 2020.

Production networks and epidemic spreading: How to restart the UK economy? https:// arxiv.org/abs/2005.10585, accessed: 07 May 2023.

National Bank of Romania 2022. Financial Stability Report. https://bnr.ro/ PublicationDocuments.aspx?icid=19966, accessed: 12 Dec 2023.

Slavata, D. 2021. Estimation of Unemployment Using Changes in the Housing Market. Public Economics and Administration 2021, 375–382.

Stein, J. C. 1995. Prices and Trading Volume in the Housing Market: A Model with Down-Payment Effects. The Quarterly Journal of Economics 110(2), 379–406.

Stock, J. H.–Watson, M. W. 2004. Combination forecasts of output growth in a seven-country data set. Journal of Forecasting 23(6), 405–430.

Tsai, I. C.–Chiang, Y. H.–Lin, S. Y. 2022. Effect of COVID-19 lockdowns on city-center and suburban housing markets: Evidence from Hangzhou, China. Journal of Asian Economic 83, 101544.

Valderrama, L.–Gorse, P.–Marinkov, M.–Topalova, P. 2023. European Housing Markets at a Turning Point. Risks, Household and Bank Vulnerabilities, and Policy Options. Washington, D.C.: International Monetary Fund.

Yilmazkuday, H. 2023. COVID-19 and housing prices: evidence from U.S. countylevel data. Review of Regional Research 43, 241–263.

Appendix 1: Net average income by county, based on averaged data. Period examined: the period between April 2020 and October 2022.

Table or figure – see the PDF version of the issue for its exact content.

Appendix 2: Construction cost index. Period examined: the period between April 2020 and October 2022

Table or figure – see the PDF version of the issue for its exact content.

Annex 3: Harmonised unemployment rate. Period under review: the period between April 2020 and October 2022

Table or figure – see the PDF version of the issue for its exact content.

Appendix 4: Interest rates of variable-rate amortising housing loans. Period examined: the period between April 2020 and October 2022

Table or figure – see the PDF version of the issue for its exact content.

Appendix 5: Economic Sentiment Index. Period examined: the period between April 2020 and October 2022

Table or figure – see the PDF version of the issue for its exact content.

Authors

  • Oborocea Mónika

A szöveget automatikusan nyertük ki a PDF-ből; a táblázatok, ábrák és a pontos tördelés a lapszám PDF-változatában láthatók.