Friday, November 18, 2011

Does Devaluation Help GDP Growth?

One of the biggest economic fallacies is the belief that devaluing a currency is the path to economic growth. This mercantilist argument is seen everywhere by all sides of the political spectrum. The argument is that by devaluing its domestic currency, a country will boost exports and reduce imports. Domestic companies will be able to better compete both at home and abroad. Thus, they will expand production and hire more people. Typically, they argue that the only downside is that travelers visiting foreign countries will be faced with higher prices.

I would like to add a few more negative consequences to the devaluation idea. First, all imports will become more expensive. And not only imports, but all exportable goods will become more expensive. Thus, even if the economy produces a surplus of oil, coal, food, cars, etc. - all of these goods can be exported. If they can be exported, and the local currency is devalued, then companies are more likely to export these goods rather than sell them in the domestic economy. This will lower the supply and thus raise the price of all exportable goods.

Higher prices for imports and exportable goods hurt the local economy in two ways. First, companies that have these goods as part of their costs will see their costs rise. If a firm's cost rise, then all else being equal, the firm will reduce production compared to what they would have done without the rise in costs. Now the devaluation supporter will argue that the firm's sales will increase, but this will not be true of all firms. Some firms such as restaurants will see their costs of food go up, and since they do not compete with foreign entities, then they will not gain any benefit from devaluation. Other firms like airlines or computer manufacturers will see costs go up and possibly sales go up. But how can we be sure that the increase in sales will offset the increase in costs? How can the supporters of devaluation know that the overall benefits will exceed the overall costs? I don't believe there is any way to know for sure. Below, however, we will see some regressions testing to see whether countries benefit from devaluation.



The second way a domestic economy is hurt from higher prices for imports and exportable goods is consumers are forced to pay higher prices. We have seen above that it is not only travelers that experience higher price, but local consumers will also face higher prices. If consumers need to pay more for food, energy, clothes, electronics, cars, houses (building material prices increase), just about everything except haircuts, then they will reduce their overall expenditures or reduce their savings. Either way, some part of the economy will take a hit - the higher order structure of production if savings is reduced and the lower order structure of production if consumption is reduced.

Now let's look at some regressions. I ran a panel regression of 131 countries with 103 currencies going back to 1991. Many countries do not go back that far, for example the Euro only goes back to 1998. The dependent variable is GDP_Growth. The independent variable is the change in gold price for that currency. I measure the extent of the devaluation by the percentage change in the price of gold for that currency. The larger the price of gold increases, the more the currency is devalued. The result of the year by year panel regression is:


GDP Growth %

Devaluation %


-4.3769e-07
(0.01105)
Adjusted R-squared: 0.0031106
N=2073

The data shows coefficient, then p-value.

The results show a very significant negative relationship between devaluation and GDP growth. Thus, the evidence supports the claim that devaluation harms economic growth rather than helps it.

However, the results economic impact is small (the low coefficient), and it has a low adjusted r-squared. I believe this is due to the fact that the economic consequences of a devaluation occur over a period of time that can extend outside of one year. Thus, I did another regression where I take 6 year averages and run a panel regression on that. Results are below:


GDP Growth % 6 yr avg

Devaluation % 6 yr avg


-0.069315
(0.0003459)
Adjusted R-squared: 0.045137
N=272

Now the coefficient shows a larger economic impact, the significance is well below 1%, and the adjusted r-squared increases to 0.045. This more clearly shows the negative impact devaluation has on a country's economic growth.

References:
http://www.imf.org/external/pubs/ft/weo/2011/02/weodata/index.aspx
http://www.oanda.com/currency/historical-rates/

1 comment:

  1. I can see two possible reasons for negative correlation between currency deval and economic growth, other than deval does not spur GDP growth. One is outlier countries like Zimbabwe. A few data points like 10000% inflation rate and -20% GDP growth can skew the regression results a lot. If you take out top and bottom 5% of countries, results may be different. The second one is about causality. A country usually devaluate it's currency after experiencing crisis or at least difficulties. And economic growth is usually slow after crisis/difficulties, even *with* currency devaluation, not *because of* currency devaluation. The economy may grow even slower without currency devaluation.

    ReplyDelete