Effective Protection

The basic formula for the effective rate of protection (ERP) on a simple one importable good using one traded input is:

where t_i = nominal tariff rate on good i, tj = nominal tariff rate on input j and a_ij = share of input j in cost of good i at free trade prices.^{[181] [182]}

The ERP measure can be extended to cover all taxes and subsidies affecting both importables and exportables.

specifically in Corden (1966), one of his best known and most cited papers.^[183] In this paper, he spells out the general equilibrium implications of the ERP concept and introduced non-traded goods explicitly into the model. The cru­cial insight concerning general equilibrium effects is that relative rather than absolute levels of ERPs matter in determining resource shifts induced by pro­tection. Also, in the presence of non-tradeables, one needs to take account of exchange rate changes in order to assess whether a particular sector is pro­tected relative to non-tradeables. The paper further discusses the “substitution problem” which arises when one drops the assumption of fixed input coeffi­cients (the a_ifi in Equation 22.1). In this case, changes in relative prices will lead to substitution between material inputs and labour and capital which, in turn, will change the input coefficients. These general equilibrium issues sur­rounding the ERP concept generated a very large literature which Corden (1971a) sought to address (see below).

Subsequent to Corden (1966, 1971a), an extensive empirical literature sprang up calculating ERPs for a large number of countries, both developed and developing. This reflected a growing concern about the costs of protec­tion and a widespread desire to boost export-led growth as a development strategy instead of relying upon import substitution. A great impetus to the measurement of ERPs in developing countries was given by the World Bank, under the leadership of Bela Balassa who popularised its measurement and use in developing strategies to boost industry and growth.^[184]

However, the heyday of ERP measurement came to an end in the late 1970s and early 1980s with the rapid development and spread of computable general equilibrium (CGE) models. These promised to provide better esti­mates of resource allocation shifts than ERPs precisely because they took into account all of the general equilibrium interactions which ERPs could not. Nevertheless, CGE models also have their own problems since they rely upon assumptions about agents’ behaviours which may often be unrealistic and empirical estimates of a large number of key parameters which are often sub­ject to wide margins of uncertainty. Despite the theoretical and empirical critiques levied against ERPs, Greenaway and Milner (2003), after an exten­sive literature survey, argue that ERPs are still a useful measurement tool to guide policy reforms of the systems of protection in developing countries.

In sum, Gorden’s nine years at Melbourne and Australian National University (ANU) were very productive. Not only did he make significant contributions to the ongoing debate on the costs of protection in Australia, but he also made a major theoretical contribution to the debate on how pro­tection should best be measured. His 1966 article provided the key theoretical foundations upon which the whole subsequent edifice of ERP measurement and use in policy analysis is based. This paper brought him to the attention of trade theorists everywhere.

However, we cannot conclude our review of Gorden’s work in Australia during this period without citing his 1965 Princeton survey of recent devel­opments in trade theory (Corden 1965). It proved to be a great hit with teach­ers and students and was much appreciated by two Nobel Prize winners in Economics, Bertil Ohlin and John Hicks (see Corden 2018: 136). Indeed, the latter’s appreciation of the survey played a role in Corden’s arrival at Oxford in 1967.

4