Patrick Minford: Responding to critiques of my article for The Critic – Free Trade is Worth Billions
Professor Patrick Minford, 18 March 2022
I am grateful to Ben Ramanauskas for taking issue with my Critic piece on the DIT’s trade modelling as it gives me the opportunity to spell out in more detail why this modelling is wrong. Fundamentally, it is to do with the evidence. Yes, ‘gravity’ or costs of distance plays a role in all trade models, as a basic element of costs.
The problem in the modelling is not this element itself alone but rather the other modelling assumptions made that are assumed to complement this gravity element, notably the ‘Armington’ assumption that goods from different countries are quite distinct and not easily substituted with each other. It is this assumption that underpins the high assumed costs of any EU border barrier and the low gains from eliminating barriers from the rest of the world. Basically, once in this ‘gravity model’ trade patterns are established, including by gravity forces, they cannot easily be altered by new trade policies.
If this was true, then the many changes in trade policies we have seen in the post-war period would have their effects well predicted by this model. We can observe the trends in trade and compare them with the trends the model predicts., side by side with what our Cardiff model predicts, based on normal ‘classical’ comparative advantage and hence without the Armington assumption. This comparison implies a statistical rejection of the DIT model, together with a clear acceptance of the ‘classical’ model. These tests are briefly described in my Civitas paper (Minford, 2022) and fully written up in Minford and Xu (2016). This work shows how the DIT model’s average predicted trends behaviour departs from history whereas the classical model’s are tolerably close. Charts in these publications show clearly how badly the gravity model predicts trade shares in particular.
The good performance of classical trade theory in explaining UK trade should not come as a surprise. UK trade theorists such as Ricardo and Adam Smith developed it to explain why the UK’s dominant free trade policies made sense. The Swedish theorists Heckscher and Ohlin elaborated it to explain how comparative advantage stems from the extent of a country’s endowments of skilled labour, land and other immobile factors of production. This has given us insight in modern times into why and with what effects on the developed world China has grown so fast on the back of rapidly growing manufacturing labour supply (Minford et al, 1997). It is a pity that this model has been abandoned by many trade economists in favour of the fashionable ‘gravity-based’ model with its Armington assumption.
Let me end by saying that while I regret errors in the DIT modelling effort, I thoroughly applaud the DIT’s efforts to negotiate a wide range of FTAs around the non-EU world. These efforts of theirs are sadly undermined by their own modelling approach.
References:
Minford,P. (2022)” Free trade under Brexit: why its benefits to the UK have been widely underestimated”, Civitas, London, Feb 2022. https://www.civitas.org.uk/publications/free-trade-under-brexit-why-its-benefits-to-the-uk-have-been-widely-underestimated/
Minford, P. and Xu, Y. (2016) “Classical or Gravity? Which Trade Model Best Matches the UK Facts” Open Economies Review, 2018, vol. 29, issue 3, No 4, 579-611
Minford,P. Riley, J. and Nowell, E. (1997)” Trade, technology and labour markets in the world economy, 1970-90: A computable general equilibrium analysis”, Journal of Development Studies, 1997, vol. 34, issue 2, 1-34. See also Corrigendum, Journal of Development Studies, 1999, 35, (6), 153-155