Национальный инвестиционный совет
| Вид материала | Доклад |
- Алена нариньяни, 22.17kb.
- Национальный Инвестиционный Совет (нис) общественное объединение, образованное в мае, 2448.69kb.
- Национальный Инвестиционный Совет (нис) общественное объединение, образованное в мае, 4911.31kb.
- Что такое неравенство?, 191.4kb.
- Национальный Фонд развития малого и среднего предпринимательства, 204.46kb.
- Нии общей реаниматологии, 142.73kb.
- Т. П. Белова национальный вопрос в социальных учениях российских конфессий, 145.86kb.
- Дисциплина: «Инвестиционный анализ», 14.98kb.
- Названия организаций, учреждений, предприятий, иностранных фирм, 218.62kb.
- Инвестиционный процесс: основные понятия и показатели, 225.4kb.
Table 3. ADF unit root test – first differences
| Name of the variable | Deterministic trend components | Lag | t-value (ADF) | Beta | Sigma | t-value (lag) | p-value (lag) | AIC |
| p_estate_cro | trend and constant | 12 | -3.915* | -2.798 | 0.08485 | -0.158 | 0.87 | -4.803 |
| p_estate_cro | constant | 11 | -4.449** | -2.848 | 0.0839 | 1.724 | 0.08 | -4.841 |
| housing_loan | trend and constant | 12 | -3.221 | 0.6652 | 0.0134 | 1.835 | 0.069 | -8.491 |
| housing_loan | constant | 12 | -2.915** | 0.637 | 0.0134 | 1.892 | 0.06 | -8.492 |
| ir_housing | trend and constant | 12 | -4.071** | -1.1705 | 0.1588 | 2.259 | 0.026 | -3.587 |
| ir_housing | constant | 12 | -3.439* | -0.635 | 0.1586 | 1.935 | 0.06 | -3.560 |
| net_wage_cro | trend and constant | 2 | -6.323** | -0.6101 | 0.0113 | -2.47 | 0.015 | -8.915 |
| net_wage_cro | constant | 3 | -4.549** | -0.2507 | 0.0115 | -0.948 | 0.345 | -8.870 |
| no_employ | trend and constant | 1 | -11.25** | -0.8311 | 0.0867 | 2.853 | 0.005 | -4.854 |
| no_employ | constant | 1 | -11.29** | -0.8291 | 0.0863 | 2.855 | 0.005 | -4.871 |
| const_vol | trend and constant | 2 | -6.109** | -0.4081 | 0.0381 | -1.58 | 0.11 | -6.488 |
| const_vol | constant | 2 | -5.983** | -0.3585 | 0.0382 | -1.761 | 0.08 | -6.493 |
| g_wage_const | trend and constant | 2 | -5.727** | -0.3676 | 0.0200 | -3.327 | 0.001 | -7.774 |
| g_wage_const | constant | 2 | -5.757** | -0.3680 | 0.0199 | -3.341 | 0.001 | -7.793 |
| for_demand | trend and constant | 2 | -5.268** | 0.3920 | 0.0552 | 2.74 | 0.007 | -5.746 |
| for_demand | constant | 2 | -5.234** | 0.4016 | 0.0551 | 2.727 | 0.007 | -5.760 |
Source: authors´ calculation.
Note: ADF - Augmented Dickey-Fuller; optimal time lag selected using Akaike information criteria; all series are seasonally adjusted and in logarithms; ** null hypothesis about existence of unit root rejected at 1 percent level, * null hypothesis about existence of unit root rejected at 5 percent level.
Results of ADF test, displayed in Table 2 and 3 confirm this conclusion. As far as test in levels are concerned, gross wage in construction is the only series for which tests results are ambiguous. On the one hand, it seems to be trend stationary, while on the other hand test specification that excludes trend indicates that gross wage in construction is I(1). Combining this information with graphical analysis we conclude that gross wage in construction is I(1). Tests in differences confirm that all series except housing loans are indeed I(1). Housing loans could be either I(1) or I(2), depending on test specification. We concluded that housing loans series is probably I(2) since in ADF test with trend, it marginally rejects “no unit root” hypothesis (5 % critical value is -2.89) and it does not reject null hypothesis for test without trend. This means that for the purposes of testing for cointegration among described series, vector autoregression model will have housing loan series in first differences, while all other series will be in levels.
Results of Johansen cointegration test are presented in Table 4. It is obvious that variables price of real estate, interest rate on housing loans, net wage, number of employed persons, volume of construction works, gross wage in construction and foreign demand form two cointegration/long run relationships at the level of significance of 1 percent, while at 10 percent significance level they form 3 relationships. Hence, no matter what significance level one chooses, in order to know the long run parameters of the relationship and corresponding adjustment coefficients, one
Table 4. Results of Johansen cointegration test
| Rank | eigenvalue | loglikelihood | trace | p-value |
| 0 | | 1750.19 | 231.18** | 0.000 |
| 1 | 0.43272 | 1783.07 | 165.42** | 0.005 |
| 2 | 0.35030 | 1808.08 | 115.40* | 0.068 |
| 3 | 0.25970 | 1825.52 | 80.514 | 0.170 |
| 4 | 0.25064 | 1842.25 | 47.043 | 0.554 |
| 5 | 0.16990 | 1853.05 | 25.443 | 0.767 |
| 6 | 0.12336 | 1860.69 | 10.170 | 0.911 |
| 7 | 0.057221 | 1864.11 | 3.3354 | 0.827 |
| 8 | 0.028344 | 1865.78 | - | - |
Source: authors´ calculation.
Note: VAR includes 3 lags selected using Akaike information criteria and F-test on significance of each lag; cointegration space contains trend; *** null hypothesis rejected at 1 percent level; ** null hypothesis rejected at 5 percent level; * null hypothesis rejected at 10 percent level.
must test the restrictions to cointegration space. Since there were eight variables and a trend restricted in the cointegration space, altogether nine restrictions have to be imposed on the space.
The following restrictions were tested:
- variable real estate prices was normalized (β p_estate_cro =1);
- housing loans, interest rate for housing loans and foreign demand are weakly exogenous with regards to other variables (α Dhousing_loan= α ir_housing= α for_demand= 0);
- in the long run real estate prices are unit elastic with regards to changes in gross wage in construction (β g_wage_const = -1);
- adjustment coefficients of average net wage and gross wage in construction are the same (α g_wage_const = α net_wage);
- adjustment coefficients of construction volume and number of employed persons are the same (α const_vol = α no_employ);
- in the long run, real estate prices is four times more elastic to changes in average net wage than to number of employed persons (β no_employ = β net_wage /4).
The restriction were jointly accepted (LR test of restrictions: Chi2(9)=1.92 [0.98]), which enabled us to obtain long run (β) and short run/adjustment (α) coefficients, presented in Table 5. At first glance it can be noticed that all the long run coefficients have the expected sign. When analyzing the coefficients, it becomes clear that domestic demand is the major driving force behind the real estate prices growth. Real estate prices are highly elastic to changes in housing loans (3.29%), while they are below unit elasticity when it comes to changes in other domestic demand variables like interest rate on housing loans (-0.17%), net wage (0.27%) and number of employed persons (0.07%). Foreign demand has only a marginal influence on domestic real estate prices, i.e. 1% increase in foreign demand increases prices by 0.05%.
The behavior of supply side factors is also expected. The rise of the volume of construction works by 1% lowers the real estate prices for 0.34%. The rise of construction costs in terms of gross wages raises the price of real estate by an equal proportion (a result of restrictions to cointegration space).
What one must take from this analysis is the fact that the real estate prices is almost ten times more elastic to changes in housing loans than to changes in volume of construction works. Hence, if housing loans continue to rise at the same pace in the future, even very intensified activity of the construction industry will be able to prevent future price hikes.
As far as adjustment coefficients are concerned, foreign demand, housing loans and interest rates on housing loans are, in line with the expectations, weakly exogenous with regards to other variables in the model. This means that in the short run they do not react to the changes in equilibrium represented by cointegration relationship. On the other hand, real estate prices have the highest adjustment coefficient, suggesting that it reacts the most after the disruption in equilibrium in order to restore the equilibrium relationship. Volume of construction works and number of employed persons share the second largest adjustment coefficient, meaning that after the disequilibrium occurs, construction activity and employment (probably employment from construction and construction related sectors like mining and quarrying) also adjusts to restore the equilibrium.
Table 5. Long run and adjustment coefficients of the cointegration vector
| Variable | Before imposing restrictions | After imposing restrictions | ||
| | β | α | β | α |
| p_estate_cro | 1.00 | -0.08 | 1.00 | -0.15 |
| ir_housing | 0.22 | 0.05 | 0.17 | 0.00 |
| Net_wage_cro | 0.67 | -0.03 | -0.27 | -0.048 |
| no_employ | -0.35 | -0.05 | -0.07 | -0.074 |
| const_vol | 0.24 | -0.05 | 0.34 | -0.074 |
| Dhousing_loan | -2.34 | -0.01 | -3.29 | 0.00 |
| g_wage_const | -1.56 | -0.03 | -1.00 | -0.048 |
| For_demand | -0.10 | -0.006 | -0.05 | 0.00 |
| Trend | 0.006 | - | 0.001 | - |
Source: authors´ calculation.
- oncluding remarks
The analysis in this paper clearly showed that domestic demand in general, and explosive growth of housing loans in particular, are the underlying determinants of real estate prices growth in Croatia, recorded in last ten years. Using the analyses, it was also determined that the real estate prices is almost ten times more elastic to changes in housing loans than to changes in volume of construction works. This means that even in the long run supply is still not strong enough to counter balance the effect of strong domestic demand driven mostly by undergoing convergence process. Here, by convergence process one assumes fast growth (or catching-up) of Croatian disposable income, convergence of Croatian lending interest rates to EU-15 averages which resulted in dramatic interest rates reductions and strong growth of financial sector that enabled dynamic lending activity373. As far as foreign demand and the involvement of Russian investors on the domestic real estate market is concerned, the model detected positive correlation between foreign demand and real estate prices, but the degree of foreign demands influence is quite limited (long run elasticity of real estate prices to foreign demand is 0.05%).
What does this result tells about the future possibilities for the Russian investors in Croatian real estate market? Since the prices are mainly driven by domestic demand (which will, due to the continuation of the convergence process, probably keep it present pace in the future) further price rises can be expected. Furthermore, since the removal of existing real estate market restrictions for foreigners is under way, one can expect large increase in foreign demand for secondary residences, while at the same time the supply response will not be elastic enough (due to legal constraints and limited capacity in construction industry) to offset the additional price increases. This would, ceteris paribus, imply faster convergence to long run equilibrium prices, but with potentially significant spillovers onto the local housing market along the way. The bottom line is: prices will probably rise further into the future. For past and future Russian investors into Croatian real estate market this means that buying real estate in Croatia was, is and will continue to be a very wise and lucrative investment.
- Literature
Abelson, P., Joyeux, R., Milunovich, G. and Chung, D. (2005). “Explaining House Prices in Australia: 1970–2003”, The Economic Record, Vol. 81., 96–103.
Abraham, J. M., and Hendershott, P. H. (1993). “Patterns and Determinants of Metropolitan House Prices, 1977–1991”, in Real Estate and the Credit Crunch (L. E. Browne and E. S.
Rosengren, Eds.), Federal Reserve Bank of Boston.
Abraham, J. M., and Hendershott, P. H. (1995). “Bubbles in Metropolitan Housing Markets”, Journal of Housing Research, no. 6.
Égert, B. and Mihaljek, D. (2007). “Determinants of House Prices in Central and Eastern Europe”, BIS Working Papers, no. 236.
Case, K. E., and Shiller, R. J. (1989). ‘‘The Efficiency of the Market for Single-Family Homes,’’ American Ecoomic Reviev, no. 79, 125–137.
Case, K. E., and Shiller, R. J. (1990). ‘‘Forecasting Prices and Excess Returns in the Housing Market,’’ American Real Estate and Urban Economics Association Journal, no. 18, 253–273.
Chinloy, P. (1992). “The Returns to Holding Housing”, Journal of Housing Economics, no. 2, 310–323.
Englund, P. and Ioannides, Y. (1997). “House Price Dynamics: An International
Empirical Perspective”, Journal of Housing Economics, no. 6, 119–136.
Hort, K. (1997). “On Price Formation and Quantity Adjustment in Swedish Housing Markets” Department of Economics, Uppsala University, Economic Studies, no. 34.
Hort, K. (1998). “The Determinants of Urban House Price Fluctuations in Sweden 1968–1994”,
Journal of Housing Economics, no. 7, 93–120.
Johansen, S. (1988). “Statistical Analysis of Cointegration Vectors”, Journal of Economic Dynamics and Control, Vol.12, 231-254;
Johansen, S. (1991). “Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregression”, Econometrica, Vol.59(6), 1551-1580;
Johansen, S. and Juselius, K. (1992). “Testing Structural Hypothesis in a Multivariate Cointegration Analysis of the PPP and UIP for UK”, Journal of Econometrics, Vol.53, 211-244.
Koskela, E., Loikkanen, H., and Vire´ n, M. (1992). “House Prices, Household Financial Market Liberalization in Finland”, European Economic Review, no. 36, 549–558.
Mankiw, N. G., and Weil, D. N. (1989). ”The Baby Boom, the Baby Bust, and the Housing Market”, Regional Science and Urban Economics, no. 19, 235–258.
Mihaljek, D. (2005). “Free Movement of Capital, the Real Estate Market and Tourism: A Blessing or a curse for Croatia on its way to European Union, in Croatian Accession to the European Union 3: Facing the Challenges of Negotiations, (Ott, K., Ed), Friedrich Ebert Stiftung, Zagreb.
Palacin, J. and Shelburne, R. (2005), “The private housing market in Eastern Europe and the CIS”, United nations Economic Commission for Europe, Discussion paper No. 6.