Sta. Ana coordinates Action for Economic Reforms. This article was published in the Opinion Section, Yellow Pad Column of BusinessWorld, August 20, 2007 edition, page S1/4.

Dani Rodrik is a favorite of economists of different stripes.  Both neo-classical and heterodox economists read and quote his papers.

In the Philippines, one fan is Raul Fabella, who recently relinquished the deanship of the University of the Philippines School of Economics.  Raul, who doesn’t mind being labeled a neo-liberal (though he isn’t one), keeps a handy diskette that is full of Rodrik’s papers. The other fans include the heterodox economists whose haven is Action for Economic Reforms.

Professor Rodrik is the rare economist who commands the respect of both the orthodox (neo-classical) and the heterodox.  He uses the neo-classical methodology. But this does not constrain him from coming out with unconventional ideas based on the evidence.

Mr. Rodrik is a heavyweight in his field. He is the first recipient of the newly created Albert Hirschman prize, given by the global Social Science Research Council.  It is most apt for Rodrik to become the first winner of the Hirschman prize.  Like Hirschman,
Rodrik is a maverick development economist.

It is not far-fetched that the prolific Rodrik bag the Nobel prize for economics. His unconventional writings on globalization, institutional change, international trade, industrial policy, and a host of development issues are on the cutting edge of economic thinking.

Many papers Rodrik has written are very relevant for the Philippines. One recent paper that merits our attention is about the exchange rate. Kudos to BusinessWorld for featuring in its 13 August 2007 issue a Rodrik piece entitled “Currency dilemmas: the value of being undervalued.”  This short article is drawn from an elaborate, scholarly research paper called, The Real Exchange Rate and Economic Growth: Theory and Evidence, July 2007.  (To download this paper, go to

In this paper Rodrik explicates how the exchange rate matters to sustain growth in developing countries.  Currency overvaluation is bad for the real sector of the economy.  More to the point, currency undervaluation is good for growth.  An undervalued currency “facilitates” growth.  He explains that undervaluation is necessary to enhance the relative profitability of tradable goods in a developing economy. The attractiveness of tradables is held back by institutional and market failures; hence the need to compensate via a competitive price.

Overvaluation damages the economy in several ways: Boom-and-bust cycles, capital account crises, serious current account deficits, corruption and rent seeking, etc.  Drawing from the Philippine experience, we can observe that an overvalued currency has been a standard reason that triggers economic crises.  For example, the peso was overvalued by about 17 percent vis-à-vis the currencies of Southeast and East Asian currencies (excluding China’s) just before the country was hit by the 1997 financial crisis.

But Rodrik goes further—not only should developing countries avoid overvaluation and its attendant instability; they should likewise welcome undervaluation to stimulate growth.

For his specification to estimate the relationship between growth (left-hand side) and undervaluation (right-hand side), Rodrik uses a data set containing 184 countries and 11 five-tear time periods from 1950-54 to 2000-04.  The specification’s independent variables are the initial income level, the terms of trade, and the country and time-period dummies.  The specification is designed to estimate “the impact of changes in undervaluation on changes in growth rates within countries.” This approach controls other growth determinants like the quality of institutions.

The regression yields a highly significant estimate, specifically for developing countries. The result also suggests that a high undervaluation translates into a considerable boost to growth.  The panel data evidence also shows that the undervaluation’s positive effect on growth is independent of time periods (pre-1980 or post-1980).  That is to say, the results are the same regardless of the changing global environment.  To quote Rodrik, “the explanation cannot be a simple export-led growth story.”

Slicing the data set to have narrower ranges for the undervaluation or overvaluation yields the same result—the co-efficient remains highly significant.

Mr. Rodrik likewise addresses the issue of causality. With regard to the real exchange rate being an endogenous (dependent) variable, Rodrik notes that in the real world, governments treat it as a policy variable.   Governments use a variety of policy interventions—fiscal, monetary and others—to influence the exchange rate.

The regression analysis also results in positive (plus sign) coefficients, which suggest that the causality’s direction is converse.  The exchange rate affects growth in this manner: A higher domestic currency price, the undervaluation, leads to higher growth. The reverse causality is unlikely—higher growth leading to a higher currency price (depreciation)—for growth attracts capital inflows that lead to currency appreciation or overvaluation.

Another way to look at the Rodrik paper is to ask: “Have those countries that managed to engineer sharp increases in economic growth done so on the back of undervalued currencies?”  Citing a 2005 paper he co-wrote with Ricardo Hausmann and Lant Prtichett, Rodrik says that growth went up by two percentage points or more, with the growth spurt sustained for eight years or more.

A sub-sample consisting of Asian countries shows an average undervaluation of more than 20 percent at the start of their growth spurt, with undervaluation still present—even increasing—during the sustained growth period.

Some examples of undervaluation in different countries across the globe and in different periods illustrate Rodrik’s analysis.  The core message from the experiences of China, India, South Korea, Taiwan, Uganda and Tanzania is that undervaluation is a growth determinant.  A sustained level of high growth, as in China, is closely tracked by Rodrik’s undervaluation index. In other country episodes, undervaluation is correlated with the countries’ growth spurts, but overvaluation or reduced undervaluation results in growth slowdown.

Nevertheless, there are cases in which depreciation leads to busts.  The Latin American and Philippine experiences come to mind.  In these cases, their currencies are in the first place overvalued, arising from heavy capital inflows (debt, portfolio investments and the like).  Their pattern of growth that is dependent on consumption, abetted by overvalued currencies, is not sustainable.  It leads to stagnation or recession.  Crises erupt, and the currencies fall.

Hence, a critical distinction has to be made between the countries that deliberately pursue undervaluation as a growth strategy and those countries that have unsustainable growth accompanied by overvalued currencies.  In the latter’s case, the onset of the bust cycle results in sudden and steep currency devaluation.

The study’s finding is clear-cut:  that an undervalued currency has a causal impact on growth.   The undervalued currency is an incentive for the tradable goods (both exports and import substitutes).  Rodrik thus poses the question “why tradables are ‘special’ from the standpoint of growth.”

He then offers two distinct sets of explanations, namely bad institutions and market failures, which impose a stiff penalty on tradables.

Bad institutions pertain to weak property rights, non-enforcement or incompleteness of contracts, corruption, and the like.  Market failures include coordination problems, learning externalities, imperfect credit markets, and wage premiums.

The institutional and market failures inflict a heavy cost on the tradable goods, as the evidence drawn from the literature shows. Other studies cited by Rodrik point out that the poor quality of institutions is a baggage shouldered by more complex, “relationship-intensive” products.  Further, tradables, relative to nontradables, are more sensitive to and thus disproportionately suffer from market imperfections.

Undervaluing the currency is therefore seen as a second-best solution to overcome the said binding constraints.  A strategy of currency depreciation creates problems. See, for example, how the deeply undervalued Chinese currency has resulted in the United States’ huge current account deficit.  Also, a sterilization of capital inflows to prevent currency appreciation entails sizeable opportunity costs.

To be sure, the ideal solution is to reform the institutions and the markets.  But as Rodrik puts it, this is like “telling developing countries that the way to get rich is to get rich.”  The second-best approach has thus become the practical option.


The Rodrik paper on the relationship of the real exchange rate and growth is a good reference point for the Philippine policy debate.  To reiterate, its main point is not only about avoiding an overvalued currency but also using an undervalued currency to stimulate and sustain growth.  In the face of institutional and market failures, undervaluation serves as an incentive for the most productive sectors, namely manufacturing and non-traditional agriculture.

It challenges the view of Philippine officialdom that favors either an appreciating peso or a market-determined exchange rate.

I recall a clarification written by Bangko Sentral ng Pilipinas (BSP) Deputy Governor Diwa Guinigundo (BusinessWorld , 7 February 2006), which was a rejoinder to my article titled Exporters’ Woes (BusinessWorld, 30 January 2006).  Although the reply was made more than a year ago, its amplification remains the BSP’s core position regarding the management of the exchange rate.

I summarize Mr. Guinigundo’s main points in his rejoinder:

  1. The BSP’s primary mandate is to keep prices stable; inflation has the greatest impact on the poor’s well being.  Hence, “the BSP should not be faulted for being ‘overly’ concerned with inflation.”  In the same vein, the BSP views monetary policy to be “best devoted to price stability.” The use of monetary policy to depreciate the peso, or for that matter to lower unemployment, “can only be done at the cost of ever-accelerating inflation.”
  2. For its inflation-targeting framework to have credibility, the BSP has to pursue a market-determined exchange rate.
  3. Structural reforms are the necessary condition for competitiveness.  Increasing productivity through structural reforms is superior to “pure price considerations” to enhance competitiveness.

The Rodrik paper actually provides an overarching answer to the above arguments.  It all begins with the government’s preferred policy framework.  If the national government subscribes to undervaluation as a policy to meet development goals, this necessarily becomes the constraint to the BSP’s actions.

The truth is, the independent BSP’s inflation targeting is constrained by the objectives set forth in the national development plan.  Starkly said, the BSP’s inflation target cannot trump the government’s development goals.

The BSP primer on inflation targeting recognizes the need to be responsive to objectives other than low inflation.  It says:  “Although the price stability objective is the BSP’s main priority, other economic goals—such as promoting financial stability and achieving broad-based, sustainable economic growth—are given consideration in policy decision-making.”

Thus, when the national government prefers a growth spurt that requires a depreciated currency, it can and should be accommodated by the BSP.  Targeting inflation or adopting a “market-determined” exchange rate strategy has to adjust and allow for undervaluation.  To quote Rodrik, “the central bank needs to signal to the public that it now cares about the real exchange rate—because the real exchange rate is important to exports, jobs, and sustainable growth. This can be done without announcing a specific target level for the exchange rate. There is a huge room to maneuver in between the extremes of targeting a specific level of the real exchange rate and disowning any interest in the real exchange rate.”

But will this sacrifice the BSP’s primary mandate to keep price stability? The BSP’s primary mandate of maintaining price stability is not disputed. After all, the strategy of undervaluation frowns upon high inflation, lest the gains from depreciation are wiped out. The question then revolves around how one defines price stability or low/moderate inflation.

A monetarist in the mode of Milton Friedman has zero tolerance for inflation.  A Keynesian has a bias for full employment, which necessitates a relaxing of the inflation target.

The tension exists between low inflation on the one hand and growth and employment on the other hand.  The role of the policymaker is to manage the tradeoff well, guided by the specific economic objectives in a given period.

Some noted economists (the Nobel laureate Joseph Stiglitz, for example) have pointed out that inflation, the low or moderate variety, does not harm output and employment.

That said, defining “low inflation” varies from country to country, taking into consideration the level of development, economic structure, history, institutions, and the like. For an advanced economy that is an inflation hawk like Germany, an inflation rate of two percent or below is the ideal standard. For developing countries, a higher but single-digit inflation rate is considered low.

In the Philippines, considering that the year-on-year inflation rate as of July 2007 is a benign 2.6 percent, the BSP has room to further relax monetary policy to address concerns regarding the appreciating peso, without violating its primary mandate.

It must likewise be stressed that inflation targeting is not an “iron-clad policy rule,” to use the term of Ben Bernanke (the current chair of the Federal Reserve System) and Frederic Mishkin.  (See their paper titled Inflation Targeting: A New Framework for Monetary Policy?, National Bureau of Economic Research Working Paper 5893, January 1997.)  The New Zealand experience, an example of the most “rule-like” inflation-targeting, shows that the Central Bank exercises the flexibility to respond to fluctuations affecting desirable objectives other than inflation like the exchange rate, employment, and output.

Peculiarly provocative—a working paper from the International Monetary Fund—is the study done by Laurence Ball and Niame Sheridan titled Does Inflation Targeting Matter? (June 2003).  The authors wanted to determine whether inflation targeting improves economic performance.  With OECD (Organisation for Economic Co-operation and Development) countries as sample, the authors found no evidence that inflation targeting made economic performance better. The authors nevertheless clarify that the results of their study are not an argument against inflation targeting.  They, for example, think that inflation targeting may have more political, rather than economic, significance.

Edmund Phelps (another recipient of the Nobel Prize in economics for his work on unemployment, wages, and inflation) has an illuminating insight into inflation targeting.  He describes himself as “a little bit friendly toward inflation targeting.”  At the same time, he is “a little bit bothered by the fact that the usual formulations of inflation targeting are kind of unrealistic.”   These formulations, says Phelps, “tend to assume that the policymaker knows all sort of things he couldn’t possibly know.” (From an article written by Rich Miller, Chicago Sun-Times, 10 October 2006.)

The recent upheaval arising from the US housing loan mortgage meltdown bears out the fact that central banks function not only to check inflation but also to intervene in fighting volatility and disturbances that affect the real sector.

Inflation targeting cannot be doctrinaire and allows what Bernanke and Mishkin call “constrained discretion.”

The law creating the BSP allows flexibility.  It states that the BSP’s primary objective is “to promote price stability conducive to a balanced and sustainable growth of the economy” (italics mine).

To be sure, it is not the BSP’s sole responsibility to fulfill the task of having a competitive exchange rate—nay, an undervalued currency.  This task should be the primary responsibility of the national government. The preference for undervaluation must come from the national government. The BSP can only take its cue from the national government.

Suppose that the BSP itself preferred a competitive exchange rate and thus made the appropriate intervention. This could prosper only in conjunction with complementary initiatives undertaken by the national government.

The gains from a currency depreciation are canceled out when government is fiscally reckless, thus canceling out the gains from depreciation.  The government has to strive for a primary fiscal surplus, an increase in revenue effort, and a rise in domestic public and private savings.

Those opposed to BSP intervention to influence the exchange rate for competitiveness argue that it is the government’s task to put in place structural reforms to raise productivity.  But this is a huge, comprehensive task that involves a long-term process.  Amidst institutional and market imperfections, the intermediate task to advance productivity and competitiveness is to have an undervalued currency that spurs the most productive sectors.

We should not be overly critical of the BSP though.  It is the national government that defines the rules of the game.  And Mrs. Gloria Arroyo, the queen, has spoken:  She wants a strong, appreciating peso.