How much can governments shift the cost of government expenditure from today’s voters to tomorrow’s generations of immigrants, without resorting to taxation that is explicitly discriminatory? I demonstrate that if their societies are absorbing continuous flows of new immigrants, we should expect governments that represent the interests of today’s population, even if that population is altruistically linked to future generations, to choose policies that shift some portion of the tax burden to the future. This bias in favor of deficit finance is not infinite. Today’s population or their descendents, together with future immigrants, ultimately pay the higher taxes necessary to finance the accumulated debt, and live with the additional excess burdens these higher taxes generate. For a given rate of immigration and policy horizon, governments balance the deadweight losses associated with fluctuating tax rates against the benefits that accrue to the initial resident population from shifting part of the burden of financing government expenditure to future immigrant families. To measure the deficit bias, I analyse the dynamic behavior of an optimal growth model with overlapping dynasties and factor taxation, calibrated for the US economy. Models with overlapping infinite-lived dynasties allow for a very clear distinction between natural population growth (an increase in the size of existing dynasties) and immigration (the addition of new dynasties). They also provide an alternative to the strict dichotomy between models with overlapping generations, where agents disregard the impact of their choices on future generations, and the quasi-Ricardian world of infinite-lived dynasties with representative agents that fully participate in both the economy and the political system in every period. The trajectory of the debt burden predicted by the model is a good match for the rise in US Federal government debt since the early 1980’s, as well as the increases in debt projected by the Congressional Budget Office over the next few decades.
