Policy commitment and credibility are important for inducing agents to make costly, irreversible investments. Policy uncertainty can delay investment and reduce the response to policy change. I provide theoretical and novel quantitative evidence for these effects by focusing on trade policy, a ubiquitous but often overlooked source of uncertainty, when a firm's cost of export market entry is sunk. While an explicit purpose of the World Trade Organization (WTO) and preferential trade agreements (PTAs) is to secure long term market access, little theoretical and empirical work analyzes the value of these agreements for reducing uncertainty to prospective exporters.

Within a dynamic model of heterogeneous firms, I show that trade policy uncertainty will delay the entry of exporters into new markets and make them less responsive to applied tariff reductions. Policy instruments that reduce or eliminate uncertainty such as PTAs or binding trade policy commitments at the WTO can increase entry even when applied protection is unchanged. I test the predictions for WTO commitments by a developed country, Australia, and the value of securing preferences through a PTA for a developing country, Portugal circa 1986.

I test the model using a disaggregated and detailed dataset of product level Australian imports in 2004 and 2006. I use the variation in tariffs and binding commitments across countries, products and time, to construct model-consistent measures of uncertainty. The estimates indicate that lower WTO commitments increase entry. Reducing trade policy uncertainty is at least as effective quantitatively as unilateral applied tariff reductions for Australia. These results illuminate and quantify an important new channel for trade creation in the world trade system.

I use Portugal's accession to the European Community (EC) in 1986 to test whether securing pre-existing preferences reduced trade policy uncertainty for firms. I use a firm-level dataset of Portugeuse exporters to show that net entry into EC was higher than what could have been achieved if the EC had simply lowered tariffs without admitting Portugal to the EC. Structural estimates from the model suggest that EC accession reduced the probabality of preference reversals to higher tariffs by up to 24 percent.