Objectives The main purpose of this research is to analyse the evolution of budgetary transparency in the Spanish Autonomous Communities (ACs) (also known as regions) and determine which socio-economic and political factors favor such evolution. To this aim, we will use the Transparency Index of the Autonomous Communities (INCAU), elaborated by Transparency International Spain, that is based fundamentally on contrasting whether these entities publish information relating to a set of 80 indicators, which try to cover the most important areas of information that the government of an Autonomous Community (AC) should make publicly available to society. In the selection of these 80 indicators, specialists and technicians from numerous institutions have been collaborating (Transparency International Spain, 2018). Specifically, INCAU has a twofold objective: (i) to evaluate the level of transparency of the governments of the different ACs and (ii) to encourage and promote the increase in the information that these institutions provide to citizens and to society as a whole. This index has been produced every two years since 2010. It only assesses whether the required information is available or not, but it does not evaluate the quality of the information published, nor that of the management of the governments of different ACs. Each AC obtains a score, the INCAU, giving rise to a classification of the level of transparency of the seventeen autonomous governments. INCAU is made up of several transparency sub-indexes: a) Information on the Autonomous Community; b) Relations with citizens and society; c) Economic-financial transparency; d) Transparency in contracts for services, works and supplies; e) Transparency in matters of land use planning, urban planning and public works; f) Indicators of the Transparency Law. In this research we will focus on the area of economic-financial transparency. Taking into account the previous literature, it has been considered that the socioeconomic factors that can explain the level of transparency are the population, the economic level (unemployment), the educational levels of citizenship, the use of the Internet by citizens, the budget deficit or surplus and the indebtedness. Political factors include the ideology of the ruling political party, the gender of the president, electoral participation and political competition. Methodology In order to measure the degree of budgetary transparency of the Spanish ACs, we use the index of economic-financial transparency published by Transparency International Spain. Therefore, the dependent variable of our model is this index, and the data have been obtained for the years 2010, 2012, 2014 and 2016. There are two possible scoring levels in each of the 80 indicators that make up the INCAU: a) 1 point: the information on the indicator is published on the AC website; b) 0 points: the information on the indicator is not published on the AC website. Based on the total score obtained by each of the ACs evaluated (the sum of their scores in the 80 indicators), Transparency International Spain prepares the final transparency ranking of the ACs (with a score ranging from 1 to 100, with 100 being the most transparent) both at an overall level and in each of the six areas of transparency that have been evaluated. For the last year considered (2016), the most transparent AC is the Basque Country, followed by the Balearic Islands, Catalonia and Madrid. On the other hand, on the less transparent side we find ACs such as Navarre, Extremadura or Aragon. In relation to the average of all the years analysed, the ACs with the best data are Catalonia, Castilla-Leon and La Rioja, with the three being above 80 points on average in the assessment of their levels of economic-financial transparency. In addition, those with the lowest levels of transparency on average in the period analysed are Castilla-la Mancha and the Comunidad Valenciana, which scarcely reach a score of 60 points. In order to analyse the explanatory factors of the levels of budgetary transparency, a linear multiple regression model has been proposed, which is estimated by means of the Ordinary Least Squares (OLS) method. Results The following table shows a summary of the hypotheses initially put forward and whether they have finally been accepted or rejected after comparing them in the model proposed in our methodology. [GRAPHICS] Conclusions/Limitations/Implications Analysis of the evolution of the ACs' economic-financial transparency index has enabled us to observe that its own publication has been one of the greatest stimuli for governments to increase their level of public information. This is a fact that undoubtedly implies a limitation to the dependent variable used in our model, since the range of variation of this is reduced over time, and many ACs are approaching to the transparency indexes of 100% in 2016. Therefore, it is possible that, over time and with the publication of new transparency indexes, the model proposed in this study will no longer be significant to the extent that the dependent variable (the transparency index) will tend to be uniform (the value of the index of all the ACs will be or will become 100%). In addition, it should not be forgotten that another of the limitations of these indicators is that the quality of the information issued by governments is not included in their analysis, but only the publication of this information itself. It should also be noted that any dissemination involves several costs, mainly in terms of staff time dedicated to the preparation of the information. However, much of the information is prepared periodically for compliance with obligations to other administrative bodies. That is why we believe that the advance of transparency in the field of autonomy is a question of predisposition, will and know-how, rather than of significant technological investments. In this sense, we also believe that regional governments should opt to continue improving their levels of transparency, especially those ACs of smaller size, as we have concluded in the study that they are the ones with lower levels of transparency. This can be achieved through improvements in the means by which they communicate information. Thus, as we have concluded, the population's access to the Internet is a factor that very positively affects the levels of transparency. Therefore, ACs must put emphasis on this type of means in order to provide quality information with sufficient clarity to the citizens. Normally, institutions tend to make data public, but not understandable to a large part of the citizenry. For this reason, the role of citizens is to continue asking for exhaustive, clear and complete information about the management of public administrations. As future areas of research that allow us to complement and deepen the study of the levels of transparency, the creation of new indexes would be positive in order to check the similarities and differences between them, including in the calculation of these indexes the quality and clarity of the information published by the public administrations. It would also be positive to analyse the evolution of the transparency index through panel data when there is a longer period of time.
