In a recent book Besley and Persson 2011 argue that fiscal capacity is strongly correlated with economic performance across countries (see also here and here). They cite important historical work by Mark Dincecco who has shown that across Europe, between 1650 and 1900, higher taxes were associated with both limited government and economic growth (see here). The following graph is from Dincecco (2011) which contains similar figures for other European countries.
This finding can be interpreted in many ways. The state capacity literature emphasizes the idea that governments need an adequate tax system in order to provide the institutional preconditions necessary for economic growth.
Perhaps there is an alternative explanation for the historical correlation between higher taxes and economic growth. This has to do with selection bias in historical data sets. Modern states did not emerge out of nowhere. They replaced pre-existing local systems of taxation, patronage, and rent seeking. We have a relatively large amount of information about what strong, central, governments were doing and what taxes they were collecting. However, we do not have much information about local regulations or tax systems that existed before the rise of modern states because these local institutions were subsumed or destroyed by the state-building process. There is plenty of evidence that these local systems imposed large deadweight losses, although it is difficult to put together a database measuring how large these distortions were (see this paper by Raphael Frank, Noel Johnson, and John Nye or just read about the Gabelle; also see Nye (1997) for this point).
The implication of this argument is that an increase in the measured size of central government need not have been associated with an increase in the total burden of government. Rather the total deadweight loss of all regulations and taxes could have gone down in the 18th and 19th centuries, even as the tax rates imposed by the central state went up.
(Note: the increase in per capita revenues in England depicted in the figure is largely driven by higher rates of taxation (notably the excise) and more effective tax collection and not by Laffer curve effects (although the growth of a market economy during the 18th century did make it easier for the state to collect taxes).