Government procurement contracts are a large part of many economies, often accounting for 15-20% of GDP. Given the significant size of these contracts and their public-sector nature, it is unsurprising that there is a long-standing tradition of protecting domestic bidders from foreign ones. A standard method of doing so has been to use a “price preference”, that is, awarding the contract to the lowest domestic bidder even if there are lower foreign bids, just so long as that domestic bid is not too much higher than the lowest foreign bid. For example, under the European Community’s rules, contracts were awarded to a member firm so long as its bid was no more than 3% higher than the lowest non-member-bid. Obviously, this is not the only method of discriminating against foreigners with tariffs being but one alternative. Despite this history, in step with the overall drive towards trade liberalization, efforts have been taken towards reducing the use of price preferences. Chief among these was the 1996 Government Procurement Agreement (GPA) which would have enforced non-discrimination in contract bidding among participating countries. Importantly, this agreement which covers a subset of WTO members primarily addresses price preferences and not other discrimination mechanisms, with tariffs still being permitted as per other WTO regulations. Thus, there is a need to understand how different methods of protectionism in procurement contract auctions compare with one another. This is especially true in the current international climate where there appears to be a marked shift towards protectionism. Comparing these policies is the goal of our analysis.
To do so, we modify a standard auction model in which two firms, one domestic and one foreign, are each endowed with a privately-known cost. Armed with this knowledge, the firms simultaneously submit expected profit-maximizing bids to the government under one of two policy settings. In the first, consistent with practice, we impose a price preference where the contract is awarded to the domestic firm so long as its bid is no more than a fixed percentage p higher than that of the foreign firm. The other is an ad valorem tariff t which is applied to the foreigner’s bid should it win. Note that this latter is equivalent to a tariff on the foreigner’s cost since there is a one-to-one mapping between winning bids and firm costs.
From the firms’ perspectives, there is an equivalence across the two policies, that is, for each price preference level p there is a tariff t = p that is equivalent both in terms of the probability of winning and expected profits. Intuitively, this happens because if the government replaces the price preference with a tariff of the same level, when the foreign firm increases its bid so that after-tariff profits are the same, this does not alter the bidding behaviour of the domestic firm, meaning that the probability of winning and expected after-tariff profits remain the same.
Turning to the government, we assume that it chooses its policy to maximize the expected sum of three things: the surplus generated from the contract (i.e. minimize the expected bid), expected tariff revenues (which may be costly to collect), and, given the inherently political nature of trade policy, the weighted profits of the domestic firm (where the weight represents the value of private profits relative to public revenues). When tariffs are costless to collect, the equivalence for firms holds for the government as well. This occurs because, even though a shift from a price preference p to a tariff of the same rate increases the expected cost of the contract (since the winning firm sets a higher bid), this increase is exactly offset by the rise in expected tariff revenue. In addition, with costless tariffs the government’s preferred policy is one that discriminates against foreign bidders, often to the detriment of global welfare (the sum of all players’ payoffs). Thus, under this condition, at its worst the GPA would have no impact on the level of protection between members in procurement contests. Further, if tariffs are constrained below the equivalence level under the WTO so that GPA signatories could not use their equivalent tariffs, the agreement would result in a more level playing field.
These results, however, assume that tariffs are costless to collect, something that runs counter to the empirical evidence. When there are, for example, enforcement and administration costs associated with tariffs, switching from a price preference to the firm-equivalent tariff results in lower expected government welfare since the foreign firm’s bid goes up by more than the after-collection cost tariff revenue. Because of this, so long as expected tariff revenues are rising in the tariff, the government finds protection less attractive when using a tariff and, if forced to abandon the price preference, it would set it such that protection is smaller than under its preferred price preference. Thus, in this setting, even if the tariff is unconstrained by WTO regulations, one would expect the introduction of the GPA to be efficiency-improving.
Together, these results suggest that the GPA, in particular with binding limits on tariffs under preferential trade agreements, can be expected to have lowered protection levels in government procurement auctions and increase global welfare. This has important lessons for the current trade negotiation situation. First, with threats to exit various trade agreements rising rapidly, our findings indicate that if countries were to quit the GPA this would result in rising protectionism with consequent welfare losses. Second, even if the GPA remains intact, if political pressures lead to rising tariffs even under the oversight of the WTO, this may serve to roll back the gains achieved by the GPA.
Branco, F., (1994); “Favoring Domestic Firms in Procurement Contracts.” Journal of International Economics, 37, 65-80.
Cole, M.T., R.B. Davies, and T. Kaplan, (2017); “Protection in Government Procurement Auctions,” Journal of International Economics, 106:134-142.
Riezman, R., and J. Slemrod, (1987); “Tariffs and Collection Costs.” Review of World Economics, 123, 545-549.
World Trade Organization, (2013); “Government Procurement.” Retrieved from http://www.wto.org/english/tratop_e/gproc_e/gproc_e.htm on July 31 2013.
 World Trade Organization (2013).
 Branco (1994).
 See WTO (2013) for details.
 For full analysis, see Cole, Davies and Kaplan (2017).
 See Riezman and Slemrod (1987).