House price determinants in Ghana: a mixed methods approach

Enviar para

Título:
House price determinants in Ghana: a mixed methods approach
Autor: BAAKO, Kingsley
Assuntos: Hedonic Modelling ; House Price Determinants ; House Price Indexation ; House Prices ; Housing Market Dynamics ; Property Valuation
Descrição: Despite its importance, research in house price index construction is limited in Ghana, leaving housing investment decisions little corroboratory evidence. Previous studies acknowledge that establishing a house price index is a crucial prerequisite to satisfactorily understanding the Ghanaian housing market in order to make meaningful interpretations of trends, robust predictions of future behaviour, as well as prudent policy interventions to strengthen the housing market. Therefore, the purpose of this research is to model house prices in Ghana, and to construct broad-based price indices for the Ghanaian housing market. Ascribing to a mixed methods research framework, this study adopts a sequential, exploratory research design typology. A qualitative phase aimed at drawing out the factors that determine house prices in Ghana was carried out through semi-structured interviews with 20 experienced, property practitioners. In total, 21 key variables were generated and discussed in detail. Several new findings emerged from this research. The qualitative results confirm existing knowledge on urban land theory which states that distance from the CBD is indirectly correlated with price, but finds that this impact is not that significant when ranked with other variables. It also uncovered interesting findings including the relevance of, and current confusion with, unexpired lease terms. In that, the interview analysis revealed that there are two disparate views regarding the impact of unexpired lease terms which are both being used in residential valuation in Ghana, possibly contributing to the wide variations in valuations. Other factors whose significant impacts are peculiar to the Ghana market include physical heterogeneity of properties and market sentiment. A follow-up survey was conducted where interviewees ranked the variables identified by relative importance. The Relative Importance Index revealed that the most impactful variables are neighbourhood security, materials used, road network and quality of workmanship; with the least impactful variables being property size, neighbourhood prestige/brand, nearness to CBD and property age. Guided by the findings of the qualitative data analysis, quantitative data was collected and analysed from the valuation reports of multiple public and private institutions. Variables collected included market value (a proxy for house price) as the dependent variable, and 11 predictor variables that capture the micro-level determinants in a total sample of 6,400 observations across 2 regions: Greater Accra and Ashanti Regions. These two regions were selected because of market depth and data availability. Also, these two regions constituted 36% of the entire nation's population, although there are 14 other regions. The empirical analysis included first constructing hedonic models for both cities, then constructing house price indices in both the local currency and USD to understand house price trends over the 10-year study period (2009 - 2018 inclusive). Of the 11 predictor variables reported, the security and utilities features are the most impactful value-infliencing variables. The magnitude of influence of these two variables (ranging between 57% and 82%) is consistent in Accra and Kumasi (the capital cities of the Greater Accra and Ashanti Regions), highlighting that in urban Ghana, these variables are of fundamental importance in value determination. This confirms the earlier qualitative results which highlights that security ranks highest among property value determinants in Ghana, with an RII of 0.98. The next most significant variable that influences property value is location, with a 16% and 27% magnitude of influence in Accra and Kumasi respectively. The impact of this variable is consistent in both the empirical and narrative literature. Additionally, macroeconomic models were constructed to investigate the macroeconomic variables that influence house prices over the 10-year period under study. The analysis found that the macroeconomic variables that influence house prices in Accra are interest rate, unemployment rate, inflation rate and urban population growth rate. For the Kumasi model, the statistically significant variables were GDP growth rate, unemployment rate and the Ghana stock index. In the combined Ghana dataset, the analysis found that the GDP growth rate and unemployment rate influence house price levels. This knowledge is useful to generate forecasts and therefore aid policy design. The models estimated explain up to 50% of the changes in house prices. Finally, the new Ghana housing index is compared to four global housing markets (United States of America, United Kingdom, Australia and South Africa). This comparison reveals that Ghana's house prices are significantly more volatile, with house prices fluctuating substantially in the middle of the 10-year period of study, while those of the comparison markets remained steady. The geometric standard deviation for Ghana's housing index over the 10-year period was 40%, while the comparison markets ranged between 3% to 6%. This calls for policy attention to stabilize the housing market. This thesis contributes to existing knowledge- theoretical and practical. The key contributions include: generating a novel dataset; creating local indices which also act as being nationally representative; adopting a new methodological approach for the Ghana market which yielded new variables; developing a data capture template and stakeholders' framework; among others. For the Ghanaian real estate sector, understanding housing price dynamics is of crucial importance to market stakeholders. This thesis provides insight on relevant microeconomic factors and how they impact house prices in Ghana. It also provides widely applicable hedonic models that can be used to estimate house prices. In particular, the construction of a broad-based house price index will allow industry practitioners to accurately assess the performance of housing as an investment asset and lend guidance to housing policy decisions at national and local levels. It also provides a much-needed source of data for further academic inquiry into housing dynamics in Ghana. Additionally, the construction of an aggregate house price index allows for performance comparisons between Ghanaian and other global housing markets. At the individual level, the index also helps buyers and sellers to understand house price trends and thus enables planning. This research is the first study in the Ghanaian context that has considered house price determinants from a qualitative point of view, and thus has significant implications for the valuation of housing in Ghana. The methods used could be applied to estimate house price indices for other developing housing markets. Source: TROVE
Data de criação/publicação: 2021
Idioma: Inglês

House price determinants in Ghana: a mixed methods approach

BAAKO, Kingsley

2021

Buscando em bases de dados remotas. Favor aguardar.