Is trade diversion bad? (2)

Now we relax our first assumption. Suppose consumption on hat is closely (but not perfectly) substituted by clothing. Notice the difference in the utility curve. In the graph 2 now we have a curly utility curve because of our assumption: both goods are closely substituted.

The story is the same as before-NJ signs FTA with WJ and utility level of North Jakarta under FTA, U_NJ^FTA, is lower than utility level under free trade, U_NJ^FT. But look at carefully, U_NJ^FTA curve crosses blue dash line T** (situation where NJ imposes tariff to both WJ and Tangerang). What does it mean? It clearly shows that bundle consumption in B gives the same utility level as bundle consumption in C. In other word, utility level under FTA is as good as utility level under tariff-to-all system.

From these two graphs, i conclude that: it is true FTA may create trade diversion. But is trade diversion really bad in terms of welfare? My answer is not necessarily. Trade diversion is not necessarily bad for the welfare of North Jakarta if West Jakarta efficiency in producing hat is quite close to Tangerang efficiency.

Some of you may argue, why I don’t draw blue dash line crossing point A which is the same as the graph 1. Well, I can do that but as in this case the individual preference is different from the one on the graph 1, the blue dash line crossing point A means nothing (why?).

What happen next if we, once again, take both assumptions off (there are substitution goods and more than one factor production). The result strongly proposes that trade diversion is not necessarily bad for welfare.

If so, does it mean that FTA is the best deal? The answer is surely no. This simple example only justifies the welfare of home country due to FTA where its partner’s efficiency in FTA is pretty close to other countries out of FTA. More importantly, in this case I assume that the cost of doing and cost of FTA itself are close to zero-which is in many cases unrealistic.

Is trade diversion bad? 1

I was involved in a debate with Pasha on Free Trade Agreement (FTA). At any rate, we share the same idea that free trade is worth advocating for. Unilateral-by reducing tariffs regardless what other countries do is the best way; multilateral-based on reciprocity principle is the most economic sensible way.

Yet, the very question here is how we go there. From here, I think, we make a different route :D. Perhaps Pasha may propose something here.


Without any disagreement between us, I suppose, FTA (or Preferential Trade Agreement you may say) is discriminatory. Moreover, we are also concerned that FTA may end in trade creation or trade diversion. Unfortunately, studies on the last issue (trade creation vs trade diversion) are still inconclusive (not only empirical but theoretical as well).


Yet, posting here interests me. My point here : “is trade diversion bad for welfare?” I’ll use an example from this posting, hopefully café’s manager and the author do not mind with this :D. For background story, please see the posting.


Suppose North Jakarta signs free trade agreement with West Jakarta and leaves Tangerang out of deal. It would be true that free trade agreement in hat bad for North-Jakarta people only and if only: first North-Jakarta people have a unique preference (there is no substitution in consuming hat) and second, hat production requires only one factor of production (which is true in this context where hat is produced only by labor).


This may look too technical. But hopefully the graphs may help us understanding the concept of trade diversion.


Look at the graph 1, red line curves are combinations of consumption bundle consisting of clothing and hat which give the same utility for North Jakarta people. U_NJ^FT means utility level of North Jakarta in free trade condition, while U_NJ^FTA is utility level under Free Trade Agreement (FTA). Green line, NJ-T, is trade pattern between NJ and Tangerang under free trade (or in technical term we may call as production possibility frontier) while dark line is pattern between NJ and West Jakarta (WJ).


In this case, only Tangerang and WJ are the only hat producers and NJ is the only clothing producer ( I assume this). Under free trade, NJ will import hat from Tangerang since Tangerang is more efficient than WJ (from the graph we can see that Tangerang produces more hats than WJ).


Now, North Jakarta and West Jakarta agree on FTA but NJ still imposes tariff on hat from Tangerang. What happens then, North Jakarta is likely to import hat from West Jakarta, even though West Jakarta is not an efficient producer. Tariff on Tangerang shifts trade pattern between Tangerang and NJ, hence the NJ-T green line shifts and becomes NJ-T* blue dash line (means that tariff reduces demand for hat from Tangerang ).


This is what trade diversion means. FTA alters consumption from the more efficient producer to less efficient one. And it reduces possibility of producing more good. At the same time we can see clearly that utility level of North Jakarta is lower now under FTA compared to free trade (just rule of thumb, as utility curve is close to origin, it means that utility level decreases-that’s why I draw line crossing the origin) Then, we compare situation between FTA and condition when North Jakarta imposes tariff to both West Jakarta and Tangerang. How can we analyze this? When NJ levies tariff from both WJ and Tangerang, NJ still imports hat from Tangerang because, even there is tariff, Tangerang is still the most efficient producer of hats.

Look at blue dash line in T** (without any loss generality, I don’t draw a new line for WJ). It is trade pattern between NJ and Tangerang when NJ imposes tariff to both countries. Under T**, Tangerang produces more hat than condition under T* (under FTA) why? because WJ’s hat is also imposed tariff by NJ. Yet remember, under T**, Tangerang produces less than under free trade, namely, T.


Under tariff for both WJ and Tangerang, NJ imports hat from Tangerang (because Tangerang is still more efficient producer than WJ). Therefore utility level exists in intersection between blue dash line T** (tariff) and green line T (free trade). In other word, utility level under tariff coincides with utility level under free trade. You may surprise why utility level under tariff is the same as utility level under free trade. This may happen if tariff revenue is returned to this kind consumer by lump-sum method. What if tariff revenue is not returned to the consumer? Then utility level under tariff lies between free trade condition and FTA. Yet the point is still the same, utility level under FTA is lower than under tariff-to-all system (Most Favored Nations tariff) and absolutely far lower than free trade.