While being civil servant does not pay much in current wages (usually), it typically pays off handsomely in other benefits: stabily of job and pay, health care benefits and especially retirement benefits. This can be viewed as a long-term contract, where the state gives less now for the the peace of mind and future incomes. Few private firms offer this kind of benefit, one likely reason is that they do not have the power to tax in the future to meet such obligations. One can thus ask whether it is a good thing for governments to offer such generous pensions. Could they do better with that money, say, invest in public infrastructure or education?
This is precisely what Gerhard Glomm, Juergen Jung, Changmin Lee and Chung Tran address in the context of emerging economies. That context tilts obviously the answer, given that these economies precisely lack in public infrastructure and education. In addition, public service wages are quite high in those countries, making generous benefits even less appropriate. Thus, there is no doubt on the result, and the interest is in the size of the effects. The authors calibrate their model to Brazil and find indeed that public investment or simple reductions in taxes would be very beneficial (on average, of course some have reasons to complain). Indeed, cutting public pensions makes civil servants save more, and this increases GDP by 4% through added capital accumulation. The rest is just gravy: reducing distorting labor income taxes increases GDP by a total of 15% or increasing public infrastructure bring it to 10%. It is, however, highly unlikely that those numbers would apply to developed economies.