On the infant industry hypothesis: the problem with that has historically been that the industries that lobby for those tariffs never stop and most of the time those short-term tariffs end up becoming permanent.
There are dangers with any interventionism, with the concentration of power encouraging all sorts of influence costs. However, in terms of the infant industry hypothesis, I'd suggest that long term inefficient protectionism is more likely with general 'strategic trade policy' (in particular where a trading bloc attempts to skew profits in their favour). Moreover, its actually difficult to find countries that haven't used protectionism to aid economic development: from free trade loving Britain to the recent tiger economies.
But, let's assume that didn't happen and the politicians actually followed through on the original plan to end the tariff- even then by imposing that tariff, you are effectively hurting every other industry in the domestic market that relies on the goods of that industry for their own production.
The infant industry hypothesis is about delivering specialisation according to comparative advantage. By definition with comparative advantage, some industries will suffer. However, that reflects efficiency criteria (with the overall result being an increase in economic activity)
In addition, the infant industry tariffs only apply to those industries which have grown enough to gain a name and lobby the government- what about all the industries that never got big enough to do that or are still only ideas written on paper?
It is the case that there will be asymmetric power between industries. However, why would they bother with the infant industry argument? A country can defend protectionism through social interests (i.e. either by reducing employment losses or by protecting low wage sectors) or through a conservative social welfare approach (i.e. where they attempt to minimise redistribution effects from trade through protectionism).
How does the government choose what industries are right to support with tariffs and which aren't? Naturally, the price mechanism would take care of it in the market, but there is no way any group of people or even modern computers can do that.
But that's the point: we have market failure and the price mechanism fails. Will government make mistakes? Certainly! Can protectionism actually reduce innovation and therefore harm long term competition? Indeed! Despite that, economic history shows considerable success by governments in the generation of economic development. At the same time, we have evidence that trade liberalisation has actually increased absolute poverty
