Another reason why proposed mercantilist policies particularly based on the currency valve option are not likely to work: different trade partners by different states or regions.
This from the Wall Street Journal Blog, (bold emphasis mine)
But the dollar’s impact will not be equal on each state or region. That’s because, for instance, Texas ships more exports to Mexico while New York sends more exports to Canada. Understandably, then, the health of Texan exporters depends more on changes in the dollar-peso rate while New York exporters care more about the U.S.-Canadian exchange.
To gauge the regional impact of exchange rates, the Federal Reserve Bank of Dallas has developed a real trade-weighted value of the dollar index for each state.
Foreign-exchange markets tend to focus on the dollar’s value versus the euro or yen. But for state exporters, the exchange rates in emerging nations and our NAFTA partners Canada and Mexico are probably more important.
“National exchange rate indexes do not always reflect individual state experiences,” the report says. “States at times face sharply different effective exchange-rate shifts, often provoked by economic or financial crises.”
This should not even be limited to the state or region level.
Competitiveness can be analytically regressed to independent enterprises where each firms operates on distinct cost structures, have different fields or areas of specialization and of the idiosyncratic competitive advantages, [even if they come from the same industry and operate in the same territory].
Where exchange rates could have diverse effects from the micro level from the different location of trading partners, the transmission mechanism of proposed exchange rate policies are likely to be ambiguous.
The other way to say this is that one size fits all exchange rate option is a political gamble undertaken by technocrats with society’s equity at stake.