Nine versions of Wagner's law are examined employing annual time-series data on Bolivia for the period 1940-2010. The analysis is an advance over previous work in several ways. First, the stationarity properties and the order of integration of the data are investigated using the Augmented Dickey-Fuller and the Phillips and Perron tests. Second, the hypothesis of a long-run relationship between different types of government expenditures and income is tested employing the methodology of cointegration analysis. Third, Error Correction Models are utilized to determine the direction of causality between the variables of interest. Lastly, the study comprises a period of seventy years, the longest of its kind for Bolivia. Consistent with Wagner's proposition, bidirectional causality is found between income and government expenditures in six of the nine versions of the law.