The ‘conventional’ view of globalisation and low tariffs, is an assumptive model based on a Ricardian* view of trade and globalisation.
However things are much less clear cut than conventional view admits. In the end, economic theory can prove nothing about whether free trade or some form of restricted trade is best from the point of view of the people living in a particular country. It can provide a framework to help to understand the consequences of alternative policies and to identify who is likely to experience gains and losses. Other ‘non-economic’ considerations may come into play as well, such as national autonomy.
Furthermore, the conventional view slides over a number of important issues concerning globalisation (and anti tariffs). Pervasive externality and informational problems are compounded by inequalities in power, particularly at the national level.
The analysis of tariff removal compares two equilibrium positions. The implicit assumption is that the economy moves instantly and costlessly from one equilibrium to the other. This is not just a simplification for theory purposes, it is common in empirical studies of changes in trade policy. These typically simulate what the economy would look like after a change in trade policy, but only consider the new equilibrium when all adjustments have taken place. However, we all know that in reality, the economy does not hop from one state to another; and in the mean time there is considerable unemployment which should not be ignore in terms of economic costs.
The economist Driskill wrote that in their enthusiasm for free trade, exponents of its benefits sometimes neglect to note that when tariffs are removed, the relative price of exportable goods must rise. People buying those goods will see their consumer surplus shrink. Whether any particular consumer is better off or worse off depends on the balance between the importable and exportable goods they buy.
Lowering protection against imports makes some people in society worse off, while others become better off. It’s not a clear cut zero sum game. We’re in danger of ignoring the welfare of our workers, the economic impact of letting steel fail, the desperate need to get our balance of payments under control and the tacit implicit cooperation or approval of China’s actions subsidising their steel making by up to 75% and flooding the world market by accepting that the low price is a natural price and a raised price (meaning also the need for firms to look closer to home for steel instead of internationally) would be “harmful”.
*The modern version of the Ricardian model assumes that there are two countries producing two goods using one factor of production, usually labour. The model is a general equilibrium model in which all markets (i.e., goods and factors) are perfectly competitive.
More information on Ricardian Model Assumptions: http://catalog.flatworldknowledge.com/bookhub/reader/28?e=fwk-61960-ch02_s03