Purpose: The purpose of this paper is to study correlations between the national real estate investment trusts (REIT) markets in the USA and the four Asia-Pacific countries of Australia, Hong Kong, Japan and Singapore, and document the extent to which the time variation present in these correlations can be explained from a set of 11 economic and financial factors. Both US dollar and local currency returns are used. Design/methodology/approach: Time-varying correlations are estimated using a DCC-GARCH model that allows for asymmetries in both the correlations and volatilities. The correlations are then regressed on a set of four economic and seven financial factors, and tests of statistical significance are conducted in order to discriminate between relevant and irrelevant explanatory variables. The authors estimate a fixed-effects panel regression as well as individual regressions for each dynamic correlation. Findings: Significant time variation is found in the four REIT correlation series. Panel regressions suggest that REIT correlations rise with increases in the interaction of national inflation rates and with higher global equity market uncertainty. It is also found that REIT correlations fall with increases in the US default risk premium and global equity market volume. Relaxing the structure imposed by the panel data model, individual regressions confirm most of the results, although there are some exceptions. It is also found that there are no substantial differences in the dynamics of the correlation coefficients when switching from the US dollar to local currency denominated returns. Practical implications: Investors in real estate securities across national markets should take into account information about the credit spread, the volatility and volume of global equity markets, and inflation rates when modeling correlations. These variables may alert the investors to the possibility that, under a set of circumstances, investing in real estate across different markets may not provide the expected diversification benefits. Another implication relates to the impact of currency hedging. It appears that the impact of switching from US dollar to local currency denominated returns does not substantially change the time dynamics of the correlations, or the importance of explanatory variables. Originality/value: Although considerable progress has been made in modelling time-varying correlations between various REIT markets, to the authors' knowledge, this is one of the first papers to investigate the underlying causes of the co-movement, especially between the US and Asia-Pacific markets. The paper's results will help investors and risk managers make better choices by identifying those factors that have more systematic effects on the change in the REIT correlations, rather than more transient forces.