Financial condition and liquidity funds in Spain: A regional approach

This paper studies the design, functioning and main drivers of the liquidity funds distributed among the Autonomous Communities by the Spanish central government since 2012. Our analysis of the institutional framework suggests that they are heterogeneous, opaque and changing instruments. Under the title of "liquidity funds", several financial instruments with quite different characteristics have been implemented. As a consequence, it is difficult to trace the financial relations between the central and regional level of government. Nevertheless, these instruments share some common characteristics that are highlighted in the paper: first, the voluntary adhesion of the regions implies that funds only accrue to claimant regions; second, beneficiary regions receive the funds under both fiscal and financial conditions, basically committing to implement budget adjustment plans and to share all relevant budgetary information with the central government. After several years of implementation, all territories but the the Basque Country and Navarre have received liquidity funds. However, not all of them have similar financial conditions. Some were in deep need of liquidity because financial markets have shut down for them, while others have simply resorted to these instruments because they imply considerably lower financial costs. Be as it may, liquidity funds have been financing around one third of regional expenditure in the last few years, and therefore it does not seem feasible that they will disappear in the short or even medium run. The paper's main goal is to extract the explanatory factors of the distribution of the liquidity funds. For that purpose, we use official statistics on the regional allocation of all liquidity instruments during 2012-2016 as our dependent variable, and several socio-economic, budgetary and institutional variables that address the different avenues through which regions could be lacking liquidity as regressors. Regarding our econometric approach, we work with a left-truncated (non-negative) continuous dependent variable, which would result in an upward bias of OLS estimations. In order to correct that bias, we implement several Tobit estimations. As we expected, we find that regional debt has a direct impact on regional liquidity needs: those most indebted territories were the ones resorting the most to the liquidity funds. On the contrary, the most dynamic regions with a higher level of GDP have used these funds to a lower extent. More interestingly, we find that regional financial needs have a demand component: population growth in the previous years shows a direct impact on liquidity needs. Interpreting this last piece of result with all due caution, it seems that the regional financing system has not been able to adjust to the fast demographic changes that took place in the last decade, with a large increase in population in several territories, while others were decreasing or stagnating.