VoxEU Column International trade

The quality of distance: Quality sorting, Alchian-Allen effect, and geography

With mega trade deals currently being negotiated, it is imperative to fully understand the various effects of trade policy. This column focuses on the implications of product quality and the type of trade costs for trade policy. Using data on agricultural prices from Japan, it argues that specific trade costs are more sensitive to geographic distance than ad valorem costs. The welfare effects of policy initiatives such as improving infrastructure may hence be larger than previously thought.

The effects of trade policy can be understood through the concept of extensive and intensive margins. With respect to extensive margins, trade policy may protect inefficient incumbent firms by increasing the costs of entering markets. Through intensive margins, trade liberalisation induces efficient firms to expand their output and raise profits. Thus, trade policy has different effects on producers within the same industry. Whether inefficient firms are characterised as high cost or low quality is important, partly because of the associated policy implications but also because quality plays a crucial role in understanding firm behaviour (Baldwin and Harrigan 2011, Khandelwal et al. 2013).

The impact of trade policy also depends on the type of trade costs. As Khandelwal et al. (2013) argue, a reduction in specific costs affects firm behaviour in a different way than an ad valorem cost reduction. If low price reflects low quality, an increase in specific costs disproportionately raises the costs of low-quality firms. Thus, a decrease in trade costs will induce entry by low-quality firms. Although it is difficult to incorporate specific costs into a general equilibrium model (Irarrazabal et al. 2013), it is necessary to incorporate heterogeneity, quality, and specific costs to represent supply and pricing patterns and consequently to understand the impact of trade policy. Therefore, it is important to address whether high prices reflect high quality and to determine the contribution of specific costs to total trade costs.

The approach taken in the literature is to examine the relationship between 'free on board' (FOB) prices and distance to market, with distance acting as a proxy for trade costs. Scholars who have obtained evidence of a positive relationship between FOB prices and distance include Bastos and Silva (2010), Baldwin and Harrigan (2011), Manova and Zhang (2012), Martin (2012), and Feenstra and Romalis (2014). This evidence motivates the introduction of quality because such a relationship is not consistent with the standard productivity heterogeneity model. In this model (Melitz 2003), because high-productivity firms can enter markets in which entry costs are high (perhaps due to high transport costs) and set low prices, there will be a negative relationship between FOB prices and distance to market. Incorporating quality into the firm heterogeneity model introduces the case in which high FOB-price firms produce high-quality goods; these firms can therefore sell to markets that are more distant.

Specific trade costs also may induce a positive relationship between distance and quality. Because the relative prices of high-quality, and therefore high-price, goods are lower in distant markets when there are specific trade costs, there is a strong relative demand for high-quality goods in these markets; this is known as the Alchian-Allen effect (Hummels and Skiba 2004). Thus, particular types of trade costs may give rise to a positive relationship. However, based on the reduced-form regressions from the literature, it is difficult to determine which of these two mechanisms is the main driving force (and indeed whether both operate).

In a recent study (Takechi 2015), I investigate the relationship between production costs and quality and the magnitude of specific trade costs. My approach has three advantages over previous studies. First, I use data on prices in place of production to measure trade costs between two regions (Anderson and van Wincoop 2004, Atkin and Donaldson 2014). Using data on Japanese wholesale agricultural product prices provides information on prices at the origin of production. Second, knowledge of origin prices and market prices can be used to determine product delivery patterns. This allows me to address the self-selection issue. Estimating trade costs by using only observed price data excludes information on undelivered items, which causes the estimation to become biased. Thus, I use a framework similar to that used by Helpman et al. (2008) and Kano et al. (2013). Third, by assuming a particular market structure, I can measure production costs and goods quality by using origin price data, because these data are unaffected by interregional trade costs. Using these implied cost and quality data enables me to examine the link between production costs and quality and how significant is the specific-cost component in trade costs.

I empirically demonstrate that high goods prices are associated with high production costs; thus, a quality heterogeneity model is appropriate. I then confirm that the presence of specific costs is significant in the trade cost function. My estimation results also reveal that the elasticity of product quality with respect to production cost is biased upward in the absence of specific trade costs. The quality elasticity is a key determinant of whether high-cost firms can make large profit by producing high-quality goods, because a higher elasticity implies greater profitability for high-cost firms. Because relative demand for high-quality goods is greater when there are specific trade costs, the presence of specific costs affects profitability in distant markets. Thus, the relationship between quality and production costs is overestimated if specific costs are ignored.

With regard to how trade costs increase with distance to market, I find that the elasticity of specific costs with respect to distance is larger than that with respect to ad valorem costs. This result is qualitatively consistent with Hummels and Skiba (2004), who assume that only specific costs depend on the distance between markets. These empirical results have implications for public policy. According to my results, because specific costs are more sensitive to geographic distance, policy initiatives such as improving infrastructure may greatly reduce specific costs by easing the burden of distance. As Irarrazabal et al. (2013) show, specific trade costs cause more distortions than ad valorem costs. Hence, welfare gains may be larger than previously thought, based on only ad valorem costs.


Anderson, J E and E van Wincoop (2004), “Trade costs,” Journal of Economic Literature 42, 691–751.

Atkin, D and D Donaldson (2014), “Who’s getting globalized? The size and implications of intranational trade costs,” mimeo.

Baldwin, R and J Harrigan (2011), “Zeros, quality, and space: trade theory and trade evidence”, American Economic Journal: Microeconomics 3, 60–88.

Bastos, P and J Silva (2010), “The quality of a firm’s exports: where you export to matters”, Journal of International Economics 82 (2), 99–111.

Feenstra, R C and J Romalis (2014), “International prices and endogenous quality”, Quarterly Journal of Economics 129, 477–527.

Helpman E, M Melitz and Y Rubinstein (2008), “Estimating trade flows: trading partners and trading volumes”, Quarterly Journal of Economics 123, 441–487.

Hummels, D., Skiba, A., 2004. “Shipping the good apples out? An empirical confirmation of the Alchian–Allen conjecture”, Journal of Political Economy 112, 1384–1402.

Irarrazabal A, A Moxnes and L D Opromolla (2013), “The tip of the iceberg: a quantitative framework for estimating trade costs”, NBER working paper 19236.

Kano K, T Kano and K Takechi (2013), “Exaggerated death of distance: revisiting distance effects on regional price dispersions”, Journal of International Economics 90, 403–413.

Khandelwal A K, P K Schott and S-J Wei (2013), “Trade liberalization and embedded institutional reform: evidence from Chinese exporters”, The American Economic Review 103(6), 2169–2195.

Manova, K., Zhang, K., 2012. “Export prices across firms and destinations”, Quarterly Journal of Economics 127, 379–436.

Martin, J (2012), “Markups, quality, and transport costs”, European Economic Review 56, 777–791.

Melitz, M J, (2003), “The impact of trade on intra-industry reallocations and aggregate industry productivity”, Econometrica 71, 1695–1725.

Takechi, K (2015), “The Quality of Distance: Quality sorting, Alchian–Allen effect, and geography”, RIETI Discussion Paper 15-E-018.

420 Reads