Abstract
Philippines: Selected Issues
Philippines: Real and Financial Cycles1
Alternative measures of the real cycle indicate that the output gap in the Philippines is near zero in 2015. Similarly, different measures of the financial cycle show that the credit gap is also near zero in 2015, with credit growth remaining below typical metrics of credit booms since the late 1990s. The interaction between real and financial cycles confirms that the cyclical position in the Philippine is broadly neutral in 2015 even if standard measures of the business cycle are augmented with financial variables.
1. The interaction between real and financial variables is crucial to assess the economy’s cyclical position. Previous studies have found strong linkages between real and financial cycles. For example, Claessens and others (2011) find that recessions associated with financial disruption episodes, notably house price busts, tend to be longer and deeper than other recessions. Conversely, recoveries associated with rapid growth in credit and house prices tend to be stronger. These findings emphasize the importance of developments in credit and housing markets for the real economy.
2. The linkages between real and financial cycles can be stronger in the presence of credit booms. Dell’Ariccia and others (2012) note that credit booms buttress investment and consumption and can contribute to long-term financial deepening. However, they often end in costly balance sheet dislocations, and, more often than not, in devastating financial crises whose cost can exceed the benefits associated with the boom. Not all booms are bad. While about a third of boom cases end up in financial crises, others do not lead to busts but are followed by extended periods of below-trend economic growth. Yet many result in permanent financial deepening which, in turn benefit long-term economic growth.
3. What does the interaction between real and financial variables imply for the current cyclical position in the Philippines? This is an important question given the acceleration in real GDP and credit growth since 2010 against the backdrop of an expansionary global financial cycle (see chapter 1). In particular, real GDP growth rose from 4.5 percent in 2000–09 to 6.2 percent in 2010–14, raising the question of whether this higher growth was due to an increase in potential growth or to cyclical factors. At the same time, credit growth accelerated from 7 percent in 2001–09 to 16.1 percent in 2010–14, giving rise to questions of whether the economy was experiencing a credit boom which, in turn was contributing to the increase in economic growth.
4. Analysis of the real and financial cycles indicates that the current cyclical position of the Philippine economy is broadly neutral. Alternative measures of the business cycle—univariate and multivariate filters and a production function approach—suggest that the output gap is near zero in 2015. Similarly, different measures of the financial cycle—deviations from trend of real credit per capita or credit-to-GDP and growth of credit-to-GDP—show that the credit gap is also near zero in 2015, with credit growth remaining below typical metrics of credit booms since the late 1990s. The interaction between real and financial cycles suggests that the cyclical position of the Philippine economy is broadly neutral in 2015, with the output gap near zero even if measures of the real cycle are augmented with financial variables.
A. Real Cycle
5. Real GDP growth in the Philippines has accelerated since 2010. From the supply side, the acceleration has been due to higher growth contributions from services and industry, partly offset by a lower growth contribution from agriculture. From the demand side, the growth pickup has been due to higher growth contributions from household and government consumption, fixed investment (equipment and construction), and changes in stocks, partly offset by a lower growth contribution from net exports of goods and services. Decomposing contributions to growth with a production function suggests that the growth pickup was mainly due to a faster TFP growth, and to a lesser extent, to faster growth of capital. How much of the increase in growth was due to potential growth vis-à-vis cyclical factors?
Contribution to Real GDP Growth, 2000–2014
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Contribution to Real GDP Growth, 2000–2014
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Contribution to Real GDP Growth, 2000–2014
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
6. Univariate filters indicate that most of the recent increase in growth has been due to higher potential growth, with the output gap near zero in 2015–16. The Hodrick-Prescott filter, the Baxter and King band-pass filter, and the Christiano and Fitzgerald symmetric band-pass filter were used to estimate potential output and the output gap in the Philippines. Actual data up to 2014 and staff projections for 2015–20 were used in the estimation. All three filters yield similar results, with potential growth increasing from about 5 percent in 2000–09 to 6½ percent in 2015–16 and output gap closing in 2015.
Potential Output and the Output Gap: Univariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Univariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Univariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
7. A multivariate filter also shows that most of the recent growth pickup has been due to higher potential growth, with the output gap near zero in 2015–16. The multivariate filter of Blagrave and others (2015) is used to estimate potential output and the output gap in the Philippines. This filter relates the output gap to slack in the labor market and inflation by including a Phillips curve and an Okun law equation into the model. Actual data at annual frequency for real GDP growth, inflation, and unemployment for 2000–14, complemented with staff’s projections for 2015–20, were used. Similar to the univariate filters above, the multivariate filter suggests that most of the recent growth pickup in the Philippines has been due to higher potential growth, which rose from about 5 percent in 2000–09 to 6½ percent in 2015–16, with the output gap around zero in 2015–16.
Potential Output and the Output Gap: Multivariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Multivariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Multivariate Filters
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
8. A production function approach also attributes most of the recent increase in growth to higher potential growth, with the output gap closing in 2015. This approach uses three factors of production: physical capital, employment, and human capital, with TFP measured as the Solow residual. Physical capital is estimated using the perpetual inventory method, while human capital is estimated interpolating the measure of educational attainment for population aged 15 and over from Barro and Lee (2013). Employment is derived from the product of working age population, labor participation rate, and employment rate (one minus the unemployment rate). In order to estimate potential output, data for TFP, labor participation, and unemployment are smoothed using the Hodrick and Prescott Filter. Physical capital, human capital, and working age population are not smoothed. The results suggest that most of the recent pickup in growth in the Philippines is explained by an increase in potential growth from below 5 percent in 2000–09 to about 6¼ percent in 2015–16. The rise in potential growth is due mainly to faster TFP growth and, to a lesser extent, to faster physical capital growth. The production function approach also shows an output gap near zero in 2015–16.
Potential Output and the Output Gap: Production Function Approach
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Production Function Approach
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Potential Output and the Output Gap: Production Function Approach
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
9. All estimates discussed above suggest that the pickup in growth in the Philippines since 2010 has been due to higher potential growth, with the output gap near zero in 2015–16. As the production function approach indicates, the acceleration in potential GDP growth has been due to higher growth of TFP and physical capital, with the contribution from labor and human capital remaining broadly stable since the late 1990s. At the same time, the pickup in growth since 2010 has not resulted in higher inflation or in a rapidly declining unemployment rate, thus leading the multivariate filter to associate the increase in growth to higher potential growth, with generally low output gaps since 2010.
B. Credit Cycle
10. Bank credit growth in the Philippines has accelerated since 2010, especially in 2014. After a long period of deleveraging following the Asian financial crisis, growth of bank loans more than doubled from 5.9 percent in 2002–09 to 15.2 percent in 2010–13 and 19.9 percent in 2014. The increase in loan growth was particularly large for services and industry. Within services, retail and wholesale trade, real estate, and financial intermediation saw the largest increases in loan growth. Within industry, manufacturing, utilities, and construction saw the largest increases. The pickup in credit growth since 2010, combined with a larger share of credit allocated to potentially speculative sectors such as real estate and construction, have raised the questions of whether the Philippines has experienced a credit boom, whether the composition of credit is a threat to financial stability, and whether the credit cycle has contributed to real sector overheating.
Contribution to Bank’s Loans Growth
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Contribution to Bank’s Loans Growth
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Contribution to Bank’s Loans Growth
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
11. Deviations of real credit per capita from trend indicate that the credit gap is slightly positive in 2015, with no signs of credit booms since the late 1990s. The approach of Mendoza and Terrones (2008) is used, which looks at deviations of real credit per capita from its Hodrick-Prescott trend, identifying credit booms when the deviation from trend is larger than 1.75 times its standard deviation. Actual data up to 2014 and staff’s projections for 2015–16 were used. This approach shows a slightly positive credit gap in 2014–16, but well below the credit boom thresholds. Moreover, this approach identifies a credit boom during 1997–98, but none afterwards regardless of the length of the sample used.
Credit Booms: Real Credit Per Capita
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Booms: Real Credit Per Capita
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Booms: Real Credit Per Capita
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
12. Deviations of credit-to-GDP from trend indicate that the current credit gap is around zero, with no credit booms since the mid 1990s, but with 2014 being a borderline case. We use the approach of Dell’Aricia and others (2012), which look at deviations of credit-to-GDP from a 10-year rolling backward-looking cubic trend. A credit boom is identified when either of the following two conditions is satisfied: (i) the deviation from trend is greater than 1.5 times its standard deviation and the growth differential between credit and nominal GDP exceeds 10 percent; or (ii) the growth differential between credit and nominal GDP exceeds 20 percent. This approach identifies credit booms in 1993 and 1995–96, when credit growth was more than 20 percentage points higher than nominal GDP growth, but finds no credit booms afterwards. It also shows that the deviation of credit-to-GDP from trend was zero in 2014, below the credit boom threshold, but with the growth differential between credit and nominal GDP close to the 10 percent threshold.
Credit Booms: Real Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Booms: Real Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Booms: Real Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
13. Changes in credit-to-GDP indicate there have been no credit booms since the late 1990s, but 2014 was a borderline case. Chapter 3 of the IMF’s Global Financial Stability Report of September 2011 finds that increases in the credit-to-GDP ratio above 3 percent, year-on-year, could serve as early warning of credit booms, with increases above 5 percent indicating more advanced and severe credit booms. For the Philippines, the changes in credit-to-GDP were above the credit boom thresholds in 1993 and 1995–97, but have been below these thresholds since (Figure 8). However, 2014 was a borderline, when the change in credit to GDP was 3.1 percent, just above the 3 percent threshold. In line with current trends, staff expects credit growth to moderate in 2015–16, with credit-to-GDP increasing by 2 percentage points per year, below the credit boom threshold.
Changes in Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.Changes in Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.Changes in Credit-to-GDP
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.14. Credit growth has moderated in the first half of 2015, and the allocation of credit has become more balanced. Bank credit growth fell from 19.9 percent (y/y) in December 2014 to 14.5 percent (y/y) in May 2015. This sharp decline in credit growth was due to a fall in overall credit during the first five months of this year, with credit growing by -0.9 percent (annualized) between December 2014 and May 2015. The slowdown in credit growth has been broad based, with almost all sectors seeing slower credit growth between May 2014 and May 2015 than during 2010–14. The allocation of credit has also become more balanced, with credit growth falling sharply for some of the sectors that saw fast credit growth during 2010–14 (construction, real estate, loans to nonresidents, wholesale and retail trade, and manufacturing). Credit growth has remained robust for the utilities sector as new power plants are being built, and for consumer loans, particularly auto loans. The recent moderation in credit growth suggests that credit activity should remain below typical metrics of credit booms in 2015–16, with lower risks to financial stability as the allocation of bank credit has become more balanced.
C. Putting It Together: Credit Neutral Output Gap
15. Financial variables contain useful information concerning the business cycle position. Expansions that coincide with rapid credit and asset price growth tend to be stronger, while recessions coinciding with credit and asset price busts tend to be longer and deeper. Borio and others (2013) argue that incorporating information about the financial cycle is important to improve measures of potential output and output gaps. Identifying potential output with non-inflationary output is too restrictive as output may be on an unsustainable path even if inflation is low and stable whenever financial imbalances are building up. Within a simple and transparent framework, they show that including information about the financial cycle can yield measures of potential output and output gaps that are not only estimated more accurately, but also more robust in real time.
Bank Credit Growth by Economic Sector
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.Bank Credit Growth by Economic Sector
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.Bank Credit Growth by Economic Sector
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Source: IMF staff estimates.16. The approach of Borio and others (2013) is used to integrate financial variables into a broader measure of the output gap. In particular, Borio and others (2013) expand a standard Hodrick-Prescott filter with data on real credit growth, and changes in real house and stock prices. This chapter uses this same set of financial variables for the case of the Philippines, but as official data on house prices does not exist, it relies upon private sector measures of condominium and land prices for the businesses districts of Metro Manila. The estimation results show that residential and land prices are not statistically significant, but growth in real stock prices is always significant. Credit growth is significant when considered alone, but it is loses its significance when real stock prices are added to the specification. Thus, two specifications were chosen, one in which only real stock prices are added to the Hodrick-Prescott filter, and another in which both real credit and real stock prices are added.
17. The results indicate that the Philippine output gap in 2015–16 is slightly larger when financial variables are included, consistent with a slightly positive credit gap. As with the approaches above, both specifications suggest that potential growth has increased from just below 5 percent in 2000–09 to just above 6 percent in 2015-16. This increase in potential growth is only slightly smaller than the one found using approaches that exclude financial variables, suggesting that most of the gains in potential growth seen in the Philippines have not been due to an overly expansionary financial cycle, but rather to other structural factors. The credit neutral approach also suggests that the output gap is slightly positive in 2015–16, consistent with a slightly positive credit/asset price gap. However, even in this case, the output gap is very small and there is significant uncertainty around these estimates. Thus, the current cyclical position in the Philippine economy seems to be broadly neutral, with the output gap near zero even if standard estimates of the output gap are augmented with financial variables.
Credit Neutral Potential Output and Output Gap
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Neutral Potential Output and Output Gap
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
Credit Neutral Potential Output and Output Gap
Citation: IMF Staff Country Reports 2015, 247; 10.5089/9781513586243.002.A002
References
Barro, Robert J. and Jong Wha Lee, 2013. “A New Data Set of Educational Attainment in the World, 1950-2010,” Journal of Development Economics, Vol 104, pp. 184–198.
Blagrave, Patrick, Roberto Garcia-Saltos, Douglas Laxton, and Fan Zhang, 2015. “A Simple Multivariate Filter for Estimating Potential Output.” IMF Working Paper No. 15/79 (Washington: International Monetary Fund).
Borio, Claudio, Piti Disyatat, and Mikael Joselius, 2013. “Rethinking Potential Output: Embedding Information About the Financial Cycle,” BIS Working Paper No. 404 (Basel: Bank for International Settlements).
Claesens, Stijn, M. Ayhan Kose, and Marco E. Terrones, 2011, “How Do Business and Financial Cycles Interact?,” IMF Working Paper No. 11/88 (Washington: International Monetary Fund).
Dell’Ariccia, Giovanni, Deniz Igan, Luc Laeven, and Hui Tong, 2012, “Policies for Macrofinancial Stability: How to Deal with Credit Booms,” IMF Staff Discussion Note No. 12/06 (Washington: International Monetary Fund).
Mendoza, Enrique, and Marco E. Terrones, 2008, “An Anatomy of Credit Booms: Evidence From Macro Aggregates and Micro Data,” IMF Working Paper No. 09/226 (Washington: International Monetary Fund).
Prepared by Jaime Guajardo and Rui Mano (both APD).
