Introduction
After the huge resurgence of interest in Purchasing Power Parity in 1982, an increasing number of economists started using new econometric methods, such as cointegration and non-stationary panel methods, to test PPP. Rogoff (1996) had introduced the so-called PPP puzzle in his paper, which concerns the question that ‘how is it possible to reconcile the extremely high short-term volatility of real exchange rates with the glacial rate (15 percent per year) at which deviations from PPP seem to die out?’ (Rogoff, 1996, p. 664). To solve the PPP puzzle, numerous explanations arose including the core of this essay, Pricing to Market. The objective of this essay is threefold: (i) to explore and review the concept of Pricing to Market (PTM), (ii) to illustrate the implications of PTM for Purchasing Power Parity, and (iii) to analyse the empirical evidence of PTM. Initially, I will start with an overview of the concept of PTM in the first part of this essay, then go on to interpret the implications of PTM for the PPP hypothesis in the following paragraph and cover the empirical evidence concerning Pricing to Market in the last section.

Main body
Pricing to Market as a concept was first introduced by Krugman in 1987 to characterise the phenomenon of imported goods’ prices staying the same or even increasing when the domestic currency appreciates. In other words, it implies that producers are capable of price discriminating among different international markets (Knetter, 1989). The fact that price discrimination for certain types of goods arise in the international goods markets may be due to the difficulty or absence of international arbitrage. Particularly, differing national standards (for instance, left-hand-drive cars are not sold in the U.K.) or monopolistic firms’ ability may both impede international goods arbitrage (Sarno and Taylor, 2002).

Apart from literal interpretation, the concept of Pricing to Market could also be illustrated using the following partial equilibrium model of exporter behaviour taken from Knetter (1989). Assume an exporting firm selling to N international destinations and demand in each destination market has the same general form:
(1)
, where s is the exchange rate (destination currency per unit of exporter’s currency), p is the price in terms of the exporting firm’s currency, v is a random variable which may impact on the demand, and qit denotes the quantity demanded by destination market i in period t. Then assume the exporter’s costs are given by:
(2) , where Ct measures costs in exporting firm’s currency units, the summation sums up all i (the destination markets), and δt ¬denotes a random (costs) shift variable in period t. The exporting firm’s profit in period t can be shown as equation (3):
(3) .
Substituting the demand functions into the profit function and maximising this with respect to the price charged in each market in each period, we can get the following set of first-order conditions:
(4)
,
Where ct equals the marginal cost of production in period t (C’ δt) and εit denotes the elasticity of demand in domestic currency price in destination market i. The system of equations in (4) captures the basic conclusion of price discrimination: the general marginal cost is equal to marginal revenue in each destination market. To be precise, ‘price in the exporter's currency is a markup over marginal cost, with the markup determined by elasticity of demand in the various destination markets’ (Knetter, 1989, p. 200).

In general, Pricing to Market makes the explanations for the Purchasing Power Parity puzzle more complete. However, the traditional explanation for the PPP puzzle is based on the presence of non-traded goods. After a large amount of evidence found from 1980s to 1990s, economists started to realize that ‘real exchange rate fluctuations are mainly attributable to failures of the law of one price among traded goods’ (Betts and Deverux, 2000, p. 216). Betts and Devereux’s (1996) model demonstrates that Pricing to Market can generate the deviations from the Purchasing Power Parity at both individual and aggregate level. That is to say, when the exporters practice Pricing to Market in the destination market, changes in exchange rate have little impact on the price of imported goods with respect to local currency (Faruqee, 1995). Thus, the deviation of Purchasing Power Parity is generated by PTM. Also, the results from the model are consistent with Krugman’s argument in 1990, suggesting that PTM is responsible for the real exchange rate volatility, dramatically. Precisely, in a large PTM sector, a depreciation of the currency does not tend to make customers spend more money on foreign goods than before. That is how the exchange rate response is magnified. Therefore, PTM takes the responsibility for the increase in exchange rate variability and the PPP puzzle as well. Basically, Pricing to Market can influence the Purchasing Power Parity in two ways. On one hand, it works as the frictions in international goods markets to impede the immediate response of domestic prices to the changes of exchange rates (Rogoff, 1996); on the other hand, it increases the volatility of real exchange rate.

The empirical findings on Pricing to Market are various and quite consistent across studies. Knetter (1989) presents a model of exporter behaviour which can permit the data to distinguish between three alternative hypotheses: the fully competitive integrated model and two other price discrimination models. In his paper, quarterly data for six U.S export products and ten German export products are considered by Knetter. And the data sample is running from 1978:1 to 1986:1 for U.S. exports and 1977:1 to 1985:4 for German exports. For the U.S. data set (breakfast cereal, dried onions, orange juice, bourdon, refrigerators and snap-action switches), Knetter finds that the country effects are statistically significant in almost all instances. Moreover, the regressions indicate that 21 export markets violate the invariance of export prices to exchange rates implied by the constant-elasticity model. Another puzzling aspect of the outcomes is that the coefficient on the exchange rate has a tendency to be more often positive than negative. This suggests that U.S. exporters actually adjust dollar prices to increase the price when an exchange rate depreciation in the foreign country occurs. This finding is consistent with Mann’s in 1986 . It is worth noting here that this optimising behaviour can be successful ‘only if exporters perceive demand schedules to be more convex than a constant elasticity of demand schedule (i.e. inelastic)’ (MacDonald, 2007, p. 82). With the German data base (wax distilled from brown coal, titanium dioxide, beer, fan belts, sparkling wine, white wine, potassium chloride, two classes of automobiles and motorcycles), Knetter finds that, in contrast to the U.S. data, the coefficient on the exchange rate term is significant in less than half of the equations. Perhaps the most interesting finding is the coefficient on exchange rate is consistently negative in every export products when the U.S. is the destination market. Knetter rules out the fixed costs of adjustment as the explanation for the finding and suggests three feasible explanations including explanation relying on market size, near competitive nature of the U.S. market and invoicing patterns. After a few years, in contrast to Knetter (1989), Knetter (1993) finds that the belief that the destination market is important in determining the extent of local-currency price stability is substantially questionable. Particularly, there is little evidence to suggest that, as a destination market, the U.S. receive remarkably different treatment from foreign exporters. Consequently, this indicates that the large swings in the dollar are not responsible for the existence of Pricing to Market. In addition, and in contrast to many previous researches in this area, little evidence of differences has been found by comparing the source-country behaviour within common industries. In the end, Knetter points out that future research should seek to explain Pricing to Market on the aspect of industry characteristics. Pick and Carter (1994) use the Knetter model and find evidence of PTM for both American and Canadian exporters. After that they extend the model with the Canadian dollar per U.S dollar exchange rate and find this exchange rate plays an important role in the export pricing decisions of both exporters. Thus, they successfully introduce the importance of competitors’ exchange rates in PTM modeling.

Marston (1990) uses monthly data on domestic prices reported at the wholesale level and FOB export prices (measured in Yen) from 1980 to 1987 for seventeen final products to study the Pricing to Market behaviour in Japanese manufacturing. The results show that fifteen products appear to have statistically significant evidence of PTM except for small trucks and cameras. Overall, the average degree of PTM found in all products is around 50%. Later, Athukorala and Menon (1994) estimate the PTM behaviour and exchange rate pass-though in Japanese exports, using quarterly data during the period 1989 to 1992. One of their findings suggests that the estimates only including the PTM effect tend to over-emphasise the degree of pass-though. Also, the most controversial outcome of their study may be the rejection of the widely accepted view that Japanese exporting firms have relied more heavily on PTM during the post-Plaza period.

Naug and Nymoen (1996) and Alexius and Vredin (1999) estimate the Pricing to Market behaviour in Norwegian and Swedish exports (both considered as small open economies). The evidence of PTM is statistically significant, and indicates that the exporters in a small open economy have the similar behaviour to those of major nations.

In the last decade, empirical findings of PTM are continuously emerging. Falk, M. and Falk, R. (2000), using panel data, find that the degree of pricing to market differs among destinations and products and the interactions between exporters are important to PTM. What is more, this finding is also supported by Bowe and Saltvedt (2004). The results from Gil-Pareja (2003) suggest that the Pricing to Market is a strong and pervasive phenomenon across industries independent of the invoicing currency. Also, Patureau (2007) improves the performances of PTM model by introducing the credit market friction.

Conclusion
In conclusion, this essay addressed Pricing to Market in three aspects: the concept, the implications for Purchasing Power Parity and the empirical evidence. Specifically, the concept of PTM was explored in the first part of this essay in two ways: literal and model interpretation. And the next paragraph stated two implications of PTM for Purchasing Power Parity: keeps the local prices sticky and increases the volatility of exchange rate. In the last part of this essay, the empirical evidence of PTM was discussed. In this part, empirical evidence in 1990s were analysed in depth (including evidence from U.S, German, Japan, Norway and Sweden) and empirical findings after 2000 were summarized.

Reference
Alexius, A. and Vredin, A. (1999) Pricing-to-market in Swedish Exports. Scandinavian Journal of Economics, 101 (2) June, pp. 223-239.

Athukorala, P. and Menon, J. (1994) Pricing to Market Behaviour and Exchange Rate Pass-Through in Japanese Exports. The Economic Journal, 104 (423) March, pp. 271-281.

Betts, C. and Devereux, M.B. (1996) The Exchange Rate in a Model of Pricing-to-Market. European Economic Review, 40 (3-5) April, pp. 1007-1021.

Betts, C. and Devereux, M.B. (2000) Exchange Rate Dynamics in a Model of Pricing-to-Market. Journal of International Economics, 50 (1) February, pp. 215-244.

Bowe, M. and Saltvedt, T.M. (2004) Currency Invoicing Practices, Exchange Rate Volatility and Pricing-to-Market: Evidence from Product Level Data. International Business Review, 13 (3) June, pp. 281-308.

Falk, M. and Falk, R. (2000) Pricing to Market of German Exporters: Evidence from Panel Data. Empirica, 27 (1), pp. 21-46.

Faruqee, H. (1995) Pricing to Market and the Real Exchange Rate. International Monetary Fund, 42 (4) December, pp. 855-881.

Gil-Pareja, S. (2003) Pricing to Market Behaviour in European Car Markets. European Economic Review, 47 (6) December, pp. 945-962.

Knetter, M.M. (1989) Price Discrimination by U.S. and German Exporters. The American Economic Review, 79 (1) March, pp. 198-210.

Knetter, M.M. (1993) International Comparisons of Pricing-to-Market Behavior. The American Economic Review, 83 (3) June, pp. 473-486.

Krugman, P.R. (1987) Pricing to Market when the Exchange Rate Changes. In: Arndt, S.W. and Richardson, J.D. ed. Real-financial Linkages among Open Economics. Cambridge, MA: MIT Press, pp. 49-70.

Krugman, P.R. (1990) Exchange-rate instability. Journal of International Economics, 28 (1-2) February, pp. 187-190.

MacDonald, R. (2007). The Economics of Exchange Rates: Theories and Evidence, Routledge, Chapter 3.

Mann, C.L. (1986) Prices, Profit Margins, and Exchange Rates. Federal Reserve Bulletin, June, pp. 366-379.

Marston, R.C. (1990) Pricing to Market in Japanese Manufacturing. Journal of International Economics, 29 (3-4) November, pp. 217-236.

Naug, B. and Nymoen, R. (1996) Pricing to Market in a Small Open Economy. The Scandinavian Journal of Economics, 98 (3) September, pp. 329-350.

Patureau, L. (2007) Pricing-to-Market, Limited Participation and Exchange Rate Dynamics. Journal of Economic Dynamics and Control, 31 (10) October, pp. 3281-3320.

Pick, D.H. and Carter, C. A. (1994) Pricing to Market with Transactions Denominated in a Common Currency. American Journal of Agricultural Economics, 76 (1) February, pp. 55-60.

Rogoff, K. (1996) The Purchasing Power Parity Puzzle. Journal of Economic Literature, 34 (2) June, pp. 647-668.

Sarno, L. and Taylor, M.P. (2002) The Economics of Exchange Rates. Cambridge. Cambridge University Press, Chapter 5.