Introduction
The global financial crisis (GFC) of 2008 shocked most countries in Central America, Panama, and the Dominican Republic (CAPDR)—much as it did emerging markets and advanced economies. Indeed, the low-growth conditions have hindered efforts to rebuild fiscal buffers and bring down high debt ratios that have hit some of the region’s economies since the financial crisis. With an expected slowdown in productivity, this chapter’s findings suggest that policymakers in most CAPDR economies should make raising potential output growth a policy priority. Structural reforms in particular must be directed at improving the investment climate, including institutional setups, and fostering more competitive and efficient product and labor markets. This will support innovation and reduce the cost of finance, both of which can bolster the medium-term prospects for economic growth.
Estimates of potential output for CAPDR countries are a particular focus of this chapter. In this context, it seeks to explain how potential output and its components—capital, labor, and their efficiency in the production process, also known as total factor productivity (TFP)—have evolved in the region since the early 2000s, and, given the likely trajectory of potential growth over the medium term, to set out the policy implications.
Before the GFC, growth in potential output was accelerating in all Central American economies (Figure 2.1). This was attributable mostly to expansion of the labor supply, increased capital accumulation, and growth in TFP, particularly in Costa Rica, the Dominican Republic, Honduras, and Panama.

Potential Output Growth (Percent)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama;DOM = Dominican Republic; CRI = Costa Rica;GTM = Guatemala; HND = Honduras; SLV = El Salvador;NIC = Nicaragua.
Potential Output Growth (Percent)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama;DOM = Dominican Republic; CRI = Costa Rica;GTM = Guatemala; HND = Honduras; SLV = El Salvador;NIC = Nicaragua.Potential Output Growth (Percent)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama;DOM = Dominican Republic; CRI = Costa Rica;GTM = Guatemala; HND = Honduras; SLV = El Salvador;NIC = Nicaragua.In the crisis aftermath, potential output growth declined in most Central American economies. Growth in potential output for the medium term is expected to continue to be slower than it was before the global financial crisis for most of the countries. This weakness appears to result from structural constraints to capital and employment growth, and the productivity slowdown. It may also reflect slow technological progress and the difficulties of developing more stable institutional, regulatory, and legal environments across the region.
Before the Crisis: How Potential Output Growth Evolved
During 2001–07, potential growth in Central American economies increased from 3.2 percent to 5.4 percent. This was exceptional. While this is partially driven by the strong performance of Panama, potential growth increased in all other economies, with the smallest improvement being in El Salvador. An acceleration in TFP can explain the bulk of the increase in Costa Rica, the Dominican Republic, Nicaragua, and Panama. In Honduras, the increase in potential growth came in response to faster capital accumulation and a small improvement in TFP, and in El Salvador and Guatemala it was employment creation and less of a drag from TFP (Figure 2.2).

Determinants of Potential Output Growth, 2001–14 (Percent, average for the period)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.
Determinants of Potential Output Growth, 2001–14 (Percent, average for the period)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.Determinants of Potential Output Growth, 2001–14 (Percent, average for the period)
Source: Garcia-Saltos and others 2016.Note: EMs = emerging market economies; PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.Total Factor Productivity Growth
TFP growth depends on technological progress and the institutional, regulatory, and legal environment in which businesses operate. TFP captures the efficiency with which labor and capital are combined to generate output, which, in turn, depends on firms’ ability to innovate, and whether the environment in which they operate fosters competition, removes unnecessary administrative burdens, provides modern and efficient infrastructure, and allows easy access to finance.
Before the crisis, most CAPDR countries experienced a significant rebound in TFP growth. TFP growth in 2006–07 increased in the years before the crisis to almost 2 percent in Costa Rica, over 3 percent in the Dominican Republic, and close to 5 percent in Panama (from close to 1 percent in 2001–03). Honduras and Nicaragua also saw some improvements in productivity growth. Possible explanations for the increase in TFP could include: shifts of resources to higher-productivity sectors, such as circuitry and mechanical parts in Costa Rica, reparation and maintenance services and high-tech equipment in Panama, and electrical equipment and medical instruments in the Dominican Republic.
Greater export diversification and economic complexity (the amount of productive knowledge that is embodied in the export content) likely contributed to the high TFP growth and long-term growth. For example, Costa Rica diversified exports to sectors characterized by technology spillovers and the upgrading of quality of products (from agricultural products and garments in the late 1980s to circuitry and mechanical parts since the mid-2000s). The Dominican Republic shifted from garments, tobacco, and mining to electrical machinery, medical instruments, and metal products. Panama and the Dominican Republic rank highest in Latin America and the Caribbean in economic complexity, and their complexity has been increasing. Indeed, the Western Hemisphere REO found that more complex and diversified economies tend to have higher GDP per capita growth (Hausmann and others 2013, IMF 2015b).
TFP growth has been absent in El Salvador and Guatemala, with some improvements in Guatemala in the years before the crisis. Productivity shortfalls in El Salvador and Guatemala, and to a lesser extent in Honduras and Nicaragua, may reflect among other factors, lags in investment in R&D and in the adoption of new technologies. Lower human capital growth (El Salvador and Honduras saw a significant decline in human capital growth from 2001 to 2007) and migration of high-skilled workers also seem to have hampered TFP growth. Productivity gains were also hindered by a lack of competition and high market concentration in various economic sectors in Central America, including agriculture. The weak business environment, including political and economic uncertainty, criminality, corruption, and poor infrastructure conditions were other factors hindering TFP growth.
Potential Employment Growth
Some Central American economies benefited from demographic dividends (further described in Chapters 2 and 8). At an early stage of the demographic transition from high fertility/mortality rates to low fertility/mortality rates, the first dividend occurs when fertility rates fall, leading to fewer young mouths to feed, and a labor force that temporarily grows faster than the population dependent on it. Other things equal, per capita income grows more rapidly too. Such a dividend is captured by a declining dependency ratio, with the largest dividend reaped by countries that experience a fall in the dependency ratio to below two-thirds. This dividend period can last five decades or more, but eventually lower fertility reduces growth of the labor force, while continuing improvements in old-age mortality increase the size of the elderly population. Other things equal, per capita income grows more slowly and the first dividend turns negative. But a second dividend is possible when a population concentrated at older working ages and skewed toward extended retirement acts to accumulate and invest assets.
Costa Rica, Panama, and the Dominican Republic benefited from demographic dividends in the 2000s (Figure 2.3). Dependency ratios in both countries had fallen below two-thirds by the 2000s, which was faster than in other economies in the region, and the working-age population share was increasing at a rapid pace. On the other hand, in El Salvador, Honduras, or Nicaragua, the dependency ratio (though steadily declining) was still above two-thirds in the early 2000s and these countries were not yet fully benefiting from demographic dividends. In Guatemala, the dependency ratio remained above two-thirds in the early 2000s. The trends in these countries highlight the need to reduce the dependency ratio to reap full benefits of demographic transition.

Demographic Indicators
Sources: UN Economic Commission for Latin America and the Caribbean (ECLAC); and WDI.Note: CRI = Costa Rica; DOM = Dominican Republic; GTM = Guatemala; HND = Honduras; NIC = Nicaragua; PAN = Panama; SLV = El Salvador.
Demographic Indicators
Sources: UN Economic Commission for Latin America and the Caribbean (ECLAC); and WDI.Note: CRI = Costa Rica; DOM = Dominican Republic; GTM = Guatemala; HND = Honduras; NIC = Nicaragua; PAN = Panama; SLV = El Salvador.Demographic Indicators
Sources: UN Economic Commission for Latin America and the Caribbean (ECLAC); and WDI.Note: CRI = Costa Rica; DOM = Dominican Republic; GTM = Guatemala; HND = Honduras; NIC = Nicaragua; PAN = Panama; SLV = El Salvador.The increase in potential growth in some countries in the 2000s mostly reflected higher labor. In El Salvador, potential employment growth increased from 1.5 percent to 1.9 percent and in Guatemala from 3.3 percent to 3.5 percent during 2001–07, mainly attributable to higher working-age population growth. In El Salvador, fertility rates were falling steeply in the 2000s, which could explain the steady increase in the working-age population and the steady decline in the dependency ratio (from 1 to 0.72 by 2005, and further below two-thirds in the late 2000s). In Guatemala too, fertility rates in the 2000s were falling while population growth was increasing due to higher life expectancy, which in part would explain the high working-age population growth at that time. Employment growth in the Dominican Republic, Honduras, and Nicaragua continued at the same rates over 2001–07 (that is, at lower rates in the Dominican Republic, where growth in the working-age population came to a halt in the 2000s, and at higher rates in Honduras and Nicaragua, where significant reductions in fertility rates were more recent, and therefore benefits from demographic dividends and increases in working-age populations took place in the 2000s).
Capital Growth
The evolution of capital deepening has been diverse across CAPDR countries. Capital growth increased most in Panama, from 3 percent to 6.8 percent, and in Honduras, from over 4.7 percent to 6.8 percent, during 2001–07, and to an important extent in Costa Rica, the Dominican Republic, and Guatemala from the mid-2000s, while it declined in Nicaragua and El Salvador (by almost 1 percentage point in El Salvador). The acceleration in capital accumulation was driven by a strong increase in the investment-to-capital ratio over 2001–07: from 7.5 percent to 11.5 percent in Panama, and from 9.2 percent to 11.5 percent in Honduras. The declines in Nicaragua and El Salvador were driven by falling investment-to-capital ratios. Capital goods imports were booming in most of these economies in the mid-2000s and in consequence physical capital was overhauled, which supported an increase in potential growth in most economies (although this was not the case in Nicaragua and El Salvador).
After the Crisis: How Potential Output Growth Declined
Significant declines in potential output growth after the global financial crisis in most Central American economies were due to declines in capital and TFP growth, and to a lesser extent potential employment growth. Potential growth declined by about 2 percentage points in Costa Rica (from 6.2 to 3.9 percent), the Dominican Republic (from 6.8 to 5.2 percent), and Honduras (from 5 to 3.2 percent) from 2006–07 to 2013–14, slightly more than 1 percentage point in El Salvador, and less than 1 percentage point in Guatemala and Panama, while it increased slightly in Nicaragua.
TFP Growth
A decline in TFP growth in many Central American economies accounted for another part of the decline in potential growth. TFP growth fell by up to 1.9 percentage points between 2006–07 and 2013–14. Lower TFP may reflect reduced investment in innovation through research and development as one of the effects of the crisis, along with continued weaknesses in the institutional, regulatory, and legal environment. Lower TFP may also reflect lower human capital growth. In the Dominican Republic, Nicaragua, and Guatemala, however, TFP growth recovered to rates achieved before the global financial crisis, and its contribution to potential growth remained over 2.5 percent in the Dominican Republic and Panama. These two countries had the highest TFP growth in the region in 2014.
Potential Employment Growth
Substantial declines in potential employment growth also contributed to the postcrisis decline in potential growth. In Costa Rica, potential employment growth fell by 0.8 percentage point between 2006–07 and 2013–14, due to a significant drop in working-age population growth, and by 0.3 percentage point in the Dominican Republic and Panama. Potential employment growth remained broadly stable in Honduras and Nicaragua, while it continued increasing after the crisis in El Salvador and Guatemala (by about 0.2 percentage point).
Capital Growth
Slower capital growth accounted for a large part of the decline in potential growth between 2006–07 and 2013–14. The slowdown in capital growth was over 2 percentage points in the Dominican Republic, Guatemala, and El Salvador, and over 3.5 percentage points in Honduras. In Panama, the expansion of the canal and other important investment projects brought about an expansion in capital growth (of 4 percentage points), and Nicaragua also experienced an increase in capital growth due to new energy projects, while capital growth in Costa Rica did not seem to have been affected by the crisis.
The collapse in aggregate demand after the global financial crisis can explain much of the decline in investment, and financial factors were an important transmission channel. For example, as credit flows start to dry up, firms may face less advantageous financing terms and tighter lending standards over an extended period. Financial crises can have a long-lasting effect on potential growth if investment-to-capital ratios are depressed for an extended time.
What is the Likely Trajectory of Potential Growth in the Medium Term?
Scenario analysis implies that potential growth in Central American economies is likely to remain below precrisis rates. Prospects for the components of potential growth—labor, capital, and TFP—are considered over the medium term (Figure 2.4). The analysis extends the historical analysis based on projected demographic patterns, and assumptions on investment-to-capital ratios that account for the long-lasting effect of financial crises on investment-to-capital ratios, which means capital growth remaining 1–3 percentage points below rates before the global financial crisis. Overall, average potential output growth is likely to remain at about 4 percent over the medium term, as shown in the figure. These scenarios are subject to significant uncertainty, as several country-specific factors could influence growth in potential output and the evolution of TFP in the medium term. Finally, these scenarios do not assume policy changes that could boost potential growth in the medium term.

Determinants of Potential Output Growth, 2015–20
(Percent, average for the medium term)
Source: Garcia-Saltos and others 2016.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.
Determinants of Potential Output Growth, 2015–20
(Percent, average for the medium term)
Source: Garcia-Saltos and others 2016.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.Determinants of Potential Output Growth, 2015–20
(Percent, average for the medium term)
Source: Garcia-Saltos and others 2016.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND = Honduras; SLV = El Salvador; NIC = Nicaragua.TFP Growth
Strong TFP growth is pivotal to sustain high GDP growth rates, but most likely the legacies of the GFC—e.g., more sluggish potential growth in advanced economies as reported in Chapter 3 of the April 2015 WEO, would make TFP growth for CAPDR countries below the levels reached before the global financial crisis. All in all, in the scenario TFP growth is projected to grow at the 2002–14 average growth rates.
Potential Employment Growth
Potential employment growth is expected to decline further in the medium term, with the biggest declines in Costa Rica, Honduras, and Nicaragua. This reflects demographic factors that are a drag on both the working-age population and trend labor force participation rates. Population growth is likely to slow in almost all Central American economies. Aging is expected to accelerate (but levels are still low in Guatemala, El Salvador, Honduras, and Nicaragua compared to the rest of Central America), lowering trend labor force participation rates, and together with slower population growth, reducing potential employment growth in the medium term.
In Costa Rica, declines in the growth of the working-age population (from an average of 2.1 percent for 2008–14 to 1.6 percent over 2015–20) and labor force participation are expected to result in a large decline in potential employment growth—the highest in the region (from 2.6 to 1.9 percent). Potential employment growth is expected to fall by 0.3–0.4 percentage point in Honduras, Nicaragua, and the Dominican Republic, due to declines in working-age population growth and labor force participation growth. In Panama, potential employment growth is expected to fall by 0.2 percentage point due to declining growth in working-age population and labor force participation. In El Salvador, the working-age population growth is expected to slow, while labor force participation growth should increase, resulting in slightly lower potential employment growth. Guatemala’s potential employment growth is not expected to change much in the medium term.
Capital Growth
Capital growth is expected to slow further from current rates. Investment-to-capital ratios have rebounded somewhat in Costa Rica and Nicaragua since 2011, while they have fallen slightly in the Dominican Republic, Honduras, and El Salvador, and have not changed much in Guatemala and Panama. However, overall, investment-to-capital ratios are likely to remain below rates reached before the global financial crisis. This is because of less favorable external financing conditions, constraints to local finance, and low returns to economic activity. If investment-to-capital ratios stay at the rates observed in 2014, capital growth will remain 1 to 3 percentage points below precrisis rates in all.
Policy Recommendations
Support innovation. Relative to emerging market economies, most Central American economies perform poorly across various facets of innovation. This includes spending on R&D, tertiary enrollment rates, number of patent applications, foreign direct investment inflows, ease of protecting investors, knowledge-intensive employment, and creative services exports. Strengthening R&D/technological diffusion will require strengthening institutions, human capital and research, and achieving higher business and market sophistication, and competition in product and labor markets. In Costa Rica, measures to increase competition include giving the Competition Commission more independence, eliminating antitrust exemptions, improving corporate governance of state-owned enterprises, and reducing barriers to entrepreneurship. In El Salvador, enhancing the capacity and clout of the Competition Authority and eliminating legal and logistical obstacles to promptly resolve the backlog of pending court cases on anticompetitive practices would help detect and punish monopolistic behavior. Adoption of the new Competition Law and the establishment of the competition agency is critical in Guatemala to reduce anticompetitive business practices and support the entry of innovative firms.
Raise the quality of human capital. Important advances in human capital are key to improving productivity. Central America lags other regions on various indicators such as mean years of schooling and students’ performance on tests (Figure 2.5), underscoring the need to improve the quality of schooling (as discussed in Chapter 7). With respect to enhancing the quality of teachers, it is critical to define teaching standards that guide the stages of teacher professionalization, establish more demanding criteria to select and retain the most talented teachers, and review the salary structure and establish incentives for professional development.

Human Capital (Mean years of schooling, 2015)
Source: Wittgenstein Center, Human Capital Data Explorer, accessible at http://dataexplorer.wittgen-steincentre.org/wcde-v2/.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND, Honduras; SLV = El Salvador; NIC = Nicaragua.
Human Capital (Mean years of schooling, 2015)
Source: Wittgenstein Center, Human Capital Data Explorer, accessible at http://dataexplorer.wittgen-steincentre.org/wcde-v2/.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND, Honduras; SLV = El Salvador; NIC = Nicaragua.Human Capital (Mean years of schooling, 2015)
Source: Wittgenstein Center, Human Capital Data Explorer, accessible at http://dataexplorer.wittgen-steincentre.org/wcde-v2/.Note: PAN = Panama; DOM = Dominican Republic; CRI = Costa Rica; GTM = Guatemala; HND, Honduras; SLV = El Salvador; NIC = Nicaragua.Transform labor markets. Addressing labor market rigidities will support creation of jobs and ultimately a higher contribution of labor to output growth. World Economic Forum-based surveys suggest certain labor market rigidities in the Dominican Republic, El Salvador, Honduras, and Nicaragua. These include inefficiencies in wage determination, the alignment of pay with productivity, the capacity to retain talent, mismatches between skills and jobs, and high informality in almost all CAPDR economies. Removing these rigidities will improve labor productivity. Facilitating access to social security systems, reducing tax distortions, and simplifying tax filing and business licensing procedures are reforms that would help reduce informality. In the Dominican Republic, reducing nonwage payroll expenses, addressing hindrances to more flexible work and conflict resolution arrangements, fostering shifts from receiving welfare to work, and promoting labor market programs would remove barriers to employment creation. In El Salvador, consideration should be given to the reform of wage bargaining by sequencing negotiations starting from tradables sectors, given that recent hikes in minimum wages disproportionately affected firms involved in tradables sectors and further increases are likely to make such firms uncompetitive.
Policies to foster higher labor force participation rates for women and to absorb returning migrants productively into the labor force would also raise employment growth. Most countries in Central America have high gender gaps and incentives for women to work—including free or subsidized childcare programs or increased children’s hours in school, providing additional time available for mothers to work—would boost female participation (as discussed in Chapter 3). The contribution of the demographic dividend to employment growth would increase if many of the young productive population decided not to migrate. Several countries suffer from weak political and socioeconomic conditions, including some of the world’s highest homicide rates and widespread gang violence, which drive ongoing migration (as discussed in Chapter 4).
Support investment. Policies should also prioritize mobilizing domestic savings to invest and build a higher capital stock. Investment-to-capital ratios are lowest in El Salvador, Guatemala, and Nicaragua, and even in Costa Rica, the Dominican Republic, and Honduras, they are lower compared to emerging markets (Figure 2.6). In addition, constraints to local finance, including entry (access to credit), borrowing (collateral requirements), or intermediation costs (such as high rate spreads resulting from information asymmetries and limited bank competition) can become bindings constraints to private investment and entrepreneurship. Countries wishing to improve returns to investment should reduce policy uncertainties and macro risks, strengthen institutions to secure property rights, and reduce red tape and corruption. Simplified entry regulations and improved security are also important. Increased public investment, and raising its efficiency, is critical to tackling infrastructure deficiencies.

Gross National Savings and Private Investment (Percent of GDP, 2013–16 average)
Source: Garcia-Saltos and others 2016.Note: EM = emerging market economies.
Gross National Savings and Private Investment (Percent of GDP, 2013–16 average)
Source: Garcia-Saltos and others 2016.Note: EM = emerging market economies.Gross National Savings and Private Investment (Percent of GDP, 2013–16 average)
Source: Garcia-Saltos and others 2016.Note: EM = emerging market economies.References
Blagrave, P., R. Garcia-Saltos, D. Laxton, and F. Zhang, 2015, “A Simple Multivariate Filter for Estimating Potential Output,” IMF Working Paper 15/79 (Washington, DC: International Monetary Fund).
Garcia-Saltos, R., I. Teodoru, and F. Zhang, 2016, “Potential Output Estimates for Central America and the Dominican Republic,” IMF Working Paper 16/250 (Washington, DC: International Monetary Fund).
Hamilton, J, 1994, Time Series Analysis, Princeton University Press, New Jersey.
Hausmann, R., C. Hidalgo, S. Bustos, M. Coscia, A. Simoes, M. Yildirim, The Atlas of Economic Complexity: Mapping Paths to Prosperity, 2013, https://mitpress.mit.edu/index.php?q=books/atlas-economic-complexity, MIT Press, Cambridge
International Monetary Fund (IMF), 2015a, “Where Are We Headed? Perspectives on Potential Output.” World Economic Outlook, April, Washington.
International Monetary Fund (IMF), 2015b, “Long-Run Growth in Latin America and the Caribbean: The Role of Economic Diversification and Complexity.” WHD Regional Economic Outlook, April, Washington.
Ljung, L., 1999, “System Identification: Theory for the User,” Prentice-Hall: New Jersey.
Okun, A. M., 1962, “Potential GNP: Its Measurement and Significance,” in Proceedings of the Business and Economic Statistics Section, pp. 98–104 (Washington: American Statistical Association).
Annex 2.1. Potential Output: A Theoretical Primer1
Potential output is generally thought of as the level of output that can be achieved without giving rise to inflation (Okun 1962). In the short term, actual output can deviate temporarily from the potential as shocks hit an economy. Deviations reflect the slow adjustment in wages and prices to shocks such as the global financial crisis, which make the reversion of output to its potential a gradual process. Slow adjustment due to “sticky” wages and prices is a key tenet of the New Keynesian macroeconomic framework used in this chapter.
The economic definition of potential output differs from the widely used concept of trend output, because it relies on an explicit framework based on economic theory. Central banks and other policy institutions typically rely on the economic definition of potential output because the underlying economic framework allows policymakers to gauge short-term trade-offs between output, inflation, and slack in the labor market.
Trend output is derived from filtering simple statistical data using various forms of moving averages or deterministic trends. This is equivalent to smoothing actual GDP over time, based on the implicit assumption that an economy is, on average, in a state of full capacity, without incorporating information from variables such as inflation or unemployment.
The definition of potential output used in this chapter is implemented empirically using multivariate filtering techniques. This approach adds economic structure to estimates by conditioning them on some basic theoretical relationships—the relationship between cyclical unemployment2 and inflation (Phillips curve) on one hand, and between cyclical unemployment and the output gap—the difference between observed and potential output—on the other (Okun’s law). One strength of this approach is that estimates of the output gap and potential are consistent with the Okun concept of potential. An example helps illustrate how the multivariate filtering framework uses the information from economic data to estimate potential. Suppose that, if at some time, actual inflation is below expectations and unemployment is above the estimated equilibrium rate, the framework will identify a situation of excess supply (a negative output gap), all else equal.
The stochastic process for output (real GDP) consists of three equations, and subject to three types of shocks:3
In the model, the output gap is defined as the deviation of real GDP, in log terms (Y), from its potential level
The level of potential output
To help identify the three output shock terms, a Phillips curve equation for inflation and equations describing the evolution of unemployment (an Okun’s law) are added. This links the evolution of the output gap (an unobservable variable) to observable data on inflation and unemployment:
Here,
After obtaining estimates of potential output and NAIRU from the multivariate Kalman filter, the analysis investigates the drivers of potential growth using a growth-accounting framework based on a standard Cobb-Douglas production function.
In sum, the multivariate filtering framework specified in this chapter strikes a balance between statistical filters, which are easily applicable to a wide range of countries, and structural models of potential output, which offer greater theoretical rigor but are difficult to construct and apply broadly. As a caveat, it should be noted that potential output is not directly observable. Therefore, the estimates are subject to statistical and model uncertainty. The latter implies that the estimates tend to vary depending on the underlying methodology.
For an in-depth discussion of the methodological aspects see, Garcia-Saltos and others 2016 and IMF 2015a.
Defined as the deviation of the unemployment rate from the structural unemployment rate or, more specifically, the nonaccelerating inflation rate of unemployment (NAIRU).
Parameter values and the variances of shock terms for these equations are estimated using Bayesian estimation techniques. More specifically, we use regularized maximum likelihood techniques (Ljung 1999). Also, Hamilton (1994) presents a general discussion of the Kalman flter, which is used to obtain estimates of the unobservable variables as part of the estimation process.