Dutch disease raised real exchange rate and caused deindustrialization 339.0

Our leading policy technocrats are near the leading edge in taking up innovative policy analysis tools for the economy – to analyze for possible Dutch disease and for a small model that accounts for the Lucas critique.

I discovered two gems from the Bangko Sentral ng Pilipinas (BSP) policy group while researching for my previous post on Innovator Peso, Technocrats quietly fortify institutions as Filipino politicos bicker

Sadly, both researches have disclaimers as personal views of the authors and, from observation in the real world, their findings do not translate beyond their source agencies silos. For example, the good proposals to cure for Dutch disease, are diluted if not changed in the implementation, I suspect, from “institutional friction”.

This post is about one of these researches on the revival of Dutch disease studies with the rise of commodity prices in the last four years and is the subject of this post.

Note: The next post will be about the BSP’s new policy analysis tool – Dynamic Stochastic General Equilibrium or DSGE as modeled for the Philippines.

Click image for link to BSP and click on SUPERFINAL layout.pmd for article.

Evidence for Dutch Disease afflicting the Philippines

Ma. Cyd N. Tuaño-Amador, Racquel A. Claveria, Ferdinand S. Co and Vic K. Delloro, henceforth called ACCD, in the research paper, Philippine Overseas Workers and Migrants’ remittances: The Dutch Disease Question … published in the Bangko Sentral Review of January 2007 concluded:

There is evidence to suggest that remittances have led to some of the symptoms of the Dutch disease phenomenon in the Philippines … we do not find the sharp decline in economic growth that some studies suggest could visit those countries that suffer from the disease.

What is Dutch disease?

Popularly, Dutch disease afflicts when massive foreign inflows from proceeds of discovered commodities like oil or minerals, foreign aid or hot money flows, or in the case of the Philippines, from OFW and migrants’ remittances, result into de-industrialization or hollowing out, change income distribution and factor allocation.

Technically, from the classic Corden and Neary (1982) paper, Dutch disease is “the coexistence within the traded goods sector of progressing and declining, or booming and lagging, sub-sectors … we are primarily concerned with the medium-run effects of asymmetric growth on resource allocation and income distribution…”

Weak evidence more a problem of methodology than reality

Being the SYNTHESiST blogger, I am free to synthesize beyond the confines of citations-based economics research. I believe the evidence is stronger for Dutch disease based on two critiques of the ACCD research methodology that I present further below.

This critiques come from persona experiences with case studies as business manager and productivity consultant to agribusiness and industry and in export processing zone (EPZ) firms in the past thirty years.

Of course, in the neoclassical perspective, these case studies are considered anecdotal and also vulnerable to attacks of generalizing from small samples. Still, being an engineer in management practice, I will share my insights for practitioners to make their own conclusions.

First critique.  The ACCD study used Bayesian inference while the original framework of Corden and Neary (1982) used an econometric model.

As such, the original study clearly specifies that the study assumes inputs from the “real economy” precisely to filter out the price difference from nominal data.

The empirical ACCD study used time-series from 2002-2007 by the nature of the chosen methodology itself. A quick review of BSP Assets and Liabilities would show that the Special Deposits Account was already sterilizing inward remittances and could thus induce weak evidence in 2006 and had PhP 385.5 billion by 2007. Thus the nominal numbers in the time series data was already affected by BSP sterilization.

Taking out the sterilization effect and adjusting data also yields other computational complications. I think I would not be wrong that, in the direction of effect, the evidence for dutch disease would be stronger without this data contamination.

Second critique.  Cordon and Neary also assumed full employment and factor mobility as is usual with neoclassical studies. [Note: This assumption that an unemployed person instantaneously finds a job is one of the boiler plate critiques of evolutionary economics on the neoclassical, that emphasizes inter-temporal and dynamic changes from Schumpeter's (1911 in German, 1934 in English ) book, The Theory of Economic Development.]

ACCD acknowledges and then dismisses the decades old structural unemployment of >7% with a footnote (#22) that I quote below:

The Corden-Neary Dutch disease model is an abstraction that rests on the simplifying assumptions discussed in footnote 9. It can be argued that the mixed findings on the presence of the Dutch disease in the Philippines is due to the absence of some of these assumptions in the Philippine setting: with an unemployment rate of 7.8 percent (July 2007), there is less than full employment in the Philippines, and with labor rigidities, labor is not perfectly mobile across sectors. Other empirical studies that test the presence of the Dutch disease also use these simplifying assumptions.

While dismissing the assumption of full employment because other empirical studies have done it, I think the ACCD study ought to have given unemployment more importance or at least identify it as an area for further study precisely because the tradeable factor in consideration here is labor – and not energy or oil as in the classic 1982 study.

Corden-Neary’s assumption of labor as fully employed and mobile between studied tradeable energy and manufacturing against non-tradeable services was solid neoclassical economics that makes price-clearing in the econometric model defensible.

The actual price of labor that accrue to the country is also dependent on the labor mix -where the trend is less domestics and more higher services. Obviously, policy initiatives like training improves that mix and increases mobility and would have created a richer but more difficult to model study with more action-oriented proposals.

Note: Another boiler plate critique of the neoclassical from the evolutionary economics is the hidden assumption that all labor are exact replacements of each other and have one price that trades instantaneously to equilibrium.

It may not be far-fetched, in the case of the Philippines OFW, the reduction of unemployment to a floor of 7% from the previous decades 10% represents the last bits of low level OFW deployment – a plug in the mobility that would tend to raise labor supply and price to the non0tradeable sector. Anecdotally, McDonald’s counter works used to be from Manila’s exclusive schools in the 1980s and had to be more expensive (because of in-house training) probinsyanos-looking workers in more recent years.

I think taking this simplifying assumption also weakened the evidence for the benefit of easier computation – a lifting of the Corden-Neary template – into an empirical study.

From my work experience, I have seen the de-industrialization and hollowing out happen much earlier than 2007. There may be other causes than Dutch disease like political uncertainty affecting risk perception, poor infrastructure, high energy cost, weak institutions like rule of law and contract, and the like.

In SYNTHESiST, my evolution towards this conclusion and the modelling issues came from Cororaton’s 1999 research on Total Factor Productivity that yield nil residuals especially in periods of high unemployment, i.e. high unemployment or law factor capacity utilization obscured the view of the residuals as black box.

ACCD Macro recommendations

Still, the ACCD study yielded a good set of recommendations that went beyond the mandate of the BSP (probably the reason for the disclaimer). I paraphrase the recommendations as follows.

The challenge with the Dutch disease therefore lies not so much on the foreign exchange coming in but
1. In directing these private inflows toward productive activities … such as entrepreneurial endeavors and investment in better health care, education and housing for the families of remitters.
2. The promotion and maintenance of macroeconomic stability as well as the implementation of structural reforms are also integral to avoiding the Dutch disease.
3. The government too can actively seek to invest in activities that would boost productivity in the non-tradeable goods sector, diversify exports and retrain and retool workers that are in the lagging export industries.
4. In particular, flexibility in product and labor markets should be enhanced so that the demand stimulus resulting from the rise in remittances do not place undue upward pressure on inflation and the real exchange rat.
5. Measures that the government is taking to foster and diversify exports, including through steps to lower the costs of doing business and to improve public infrastructures, are moves in the right direction.
6. Finally, there is no substitute for prudent macroeconomic policies.

What the Bangko Sentral adopted

The BSP adopted only recommendation #2 that is within its institutional mandate – instituting the Special Deposit Account (SDA) to sterilize the purchase of inward foreign currency remittance to minimize monetary liquidity and avoid inflation.

As the recorded remittance level grew through the formal channels, the pressure of higher nominal rates increase so that the sterilization became a reserve losing proposition especially when the secular trend of remittance moves in pro-cycle with the weakening US$.

I think the capital account of the BSP was diminished by November 2011 hence the need for a long-delayed recapitalization.

Final words on coping with Dutch disease

The government and country is well-served to consider the other macro-prudential proposals in this study – especially if they agree with my critique that the evidence of the Dutch disease is stronger than the ACCD empirical study concluded.

This is a greater challenge that must be done on the economy even as massive changes in the political culture are being implemented.

Beyond improving national value-adding, it also implies an improvement in government productivity not just in first-order cost reduction but, more importantly, in time-based and qualitative improvement in coordination between the legacy bureaucratic silos.

References:

Amador, C. (2004, July). Migrant remittances: Macroeconomic implications and policy prescriptions. Bangko Sentral Review.
Amador, C., Claveria, R., Co, F., & Delloro, V. (2007, January). Philippine Overseas Workers and Migrants’ Remittances: The Dutch Disease Question and the Cyclicality Issue. Bangko Sentral Review.
Bangko Sentral ng Pilipinas. (n.d.). 43 – BSP Assets and Liabilities. Retrieved December 13, 2011, from bsp.gov.ph/statistics. http://www.bsp.gov.ph/statistics/spei-new/tab43.htm
Beduya, M. (2009, March 9). Cororaton Says the Philippines Makes Nil Total Factor Productivity. Retrieved December 14, 2011, from SYNTHESiST: http://synthesistblog.com
Beduya, M. (2011, November 27). Technocats quietly fortify institutions as Filipino politicos bicker. Retrieved December 12, 2011, from SYNTHESiST: http://synthesistblog.com
Corden, W., & Neary, J. (1982). Booming Sector and De-industrialization in a Small Open Economy. The Economic Journal, 92 (368), 825-848.
Lartey, E., Mandelman, F., & Acosta, P. (2009). Remittances, Exchange Rate Regimes and the Dutch Disease: A panel data analysis. BSP International Research Conference on Remittances. Bangko Sentral ng Pilipinas.