Abstract
Selected Issues
Asset Quality and Economic Activity1
Bank asset quality can have important effects on economic activity. This Selected Issues Paper reviews recent developments in asset quality in Panama and the potential importance of asset quality as a driver of macro-financial feedback loops. The results suggest that the recent deterioration in asset quality in Panama could affect bank lending and contribute to adverse macro-financial feedback loops. Asset quality should be closely monitored to prevent the emergence of systemic risks from the build-up of debt in recent years, particularly in the context of rising interest rates.
A. Introduction
1. Bank asset quality can have important effects for both financial stability and economic activity. While Panamanian bank asset quality remains generally sound, asset quality has deteriorated in recent years following a period of robust credit growth in an environment of low global interest rates. With asset quality expected to come under pressure in the context of rising global interest rates in Panama’s predominantly variable-rate environment, this paper considers the potential importance of asset quality as a driver of macro-financial feedback loops in Panama to understand the financial stability and macroeconomic challenges that could be triggered by a further and persistent deterioration in asset quality.
B. Taking Stock of Panamanian Banks’ Asset Quality
2. Panamanian bank asset quality remains sound. While bank asset quality has deteriorated, with the level of NPLs doubling since 2015, NPLs remain relatively low compared to regional peers at 1.8 percent of total loans as of June 2018 for the national banking system and appear adequately provisioned (Figure 1). The deterioration in asset quality has, however, been widespread across the banking system, across official banks, private Panamanian banks and foreign banks operating in the onshore banking system as well as offshore banks. While there is considerable heterogeneity in the ratio of NPLs across banks, the distribution has also shifted in recent years, with more banks experiencing higher NPL ratios. By sector, commerce (which includes services), mortgages, construction and personal loans have seen the largest uptick in NPLs (commerce, mortgages, construction and personal loans) and also account for the bulk of the stock of NPLs.
Panama: Asset Quality of Panamanian Banks
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
Panama: Asset Quality of Panamanian Banks
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
Panama: Asset Quality of Panamanian Banks
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
C. Macro-Financial Implications of Asset Quality
3. High levels of NPLs affect bank lending and may result in adverse macro-financial feedback loops. High NPLs typically reduce the supply of credit, including by reducing bank profitability, tying up capital because of higher risk weights on impaired assets, and raising banks’ funding costs because of lower expected revenue streams and investors’ heightened risk perceptions (Figure 2). In turn, reduced credit supply, can contribute to weaker economic activity, with adverse implications for NPLs. Indeed, the NPL ratio in Panama is negatively correlated with growth in credit to the private sector. Growth in credit to the private sector is positively correlated with economic activity, suggesting that high NPLs are associated with subdued growth and rising unemployment.
Panama: NPLs, Bank Profitability, Credit and Economic Activity
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
Panama: NPLs, Bank Profitability, Credit and Economic Activity
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
Panama: NPLs, Bank Profitability, Credit and Economic Activity
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
4. Feedback effects between asset quality in the banking sector and the real economy are assessed using a vector-autoregression (VAR) model. The model includes the NPL ratio, growth in credit to the private sector (year-over-year), real GDP growth (year-over-year), total (headline) consumer price inflation (year-over-year), and the commercial interest rate.2,3,4 All variables are considered endogenous in the estimation of the VAR. Macro-financial feedback effects are assessed using orthogonalized impulse response functions, which illustrate the behavior of one variable in response to innovations in another variable, holding other shocks constant.5 Data constraints restrict the estimation period of the VAR to 2003Q3 to 2018Q1. Such a limited sample period significantly hinders the empirical assessment of macro-financial linkages in Panama as the economy has undergone a sustained period of high economic growth over this period, with little cyclical variability. Nevertheless, simple correlations of key macro-financial variables suggest that asset quality is persistent and negatively correlated with credit growth, economic growth, and inflation, and positive correlated with interest rates.6
5. VAR estimates confirm the presence of strong macro-financial linkages in Panama. A deterioration in bank asset quality, or a positive shock to the NPL ratio, has a statistically significant negative impact on growth in credit to the private sector, real GDP growth, and inflation (Figure 3). At its peak, a 1 percentage point increase in the NPL ratio reduces credit growth by 1.5 percentage points. Negative implications for credit growth are also persistent, with the deterioration in asset quality negatively affecting credit for longer than a year. Economic activity also declines following a deterioration in bank asset quality – the peak impact on growth of a 1 percentage point increase in the NPL ratio is estimated at 1.2 percentage points. Conversely, macroeconomic performance also has a significant effect on asset quality: a stronger macroeconomic environment improves borrowers’ debt servicing capacity and leads to a statistically significant decline in the NPL ratio.7 Panel VAR variance decompositions (Table 2) indicate that about 20 and 10 percent of the variance in growth is explained by shocks to the NPL ratio over 4 and 8-quarter horizons, respectively. The impact of growth on asset quality is smaller, with shocks to growth explaining about 10 percent of the variance in the NPL ratio over both 4- and 8-quarter horizons.
Panama: Macrofinancial Linkages 1/
(impact of 1 standard deviation increase in the NPL ratio)
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Panama: Macrofinancial Linkages 1/
(impact of 1 standard deviation increase in the NPL ratio)
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Panama: Macrofinancial Linkages 1/
(impact of 1 standard deviation increase in the NPL ratio)
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Panama: Contemporaneous and Lagged Correlations Across Macro-Financial Variables 1/
(quarterly frequency)
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
Significance at 10 percent.
Panama: Contemporaneous and Lagged Correlations Across Macro-Financial Variables 1/
(quarterly frequency)
| NPL | Credit | GDP | CPI | Interest rate | NPL (t-1) | Credit (t-1) | GDP (t-1) | CPI (t-1) | Interest rate (t-1) | |
|---|---|---|---|---|---|---|---|---|---|---|
| NPL | 1.0000 | |||||||||
| Credit | -0.2349* | 1.0000 | ||||||||
| GDP | -0.0672 | 0.6887* | 1.0000 | |||||||
| CPI | -0.2326* | 0.6491* | 0.4799* | 1.0000 | ||||||
| Interest rate | 0.1994 | -0.2778* | -0.2064 | -0.3584* | 1.0000 | |||||
| NPL (t-1) | 0.9513* | -0.1897 | 0.0091 | -0.2058 | 0.1401 | 1.0000 | ||||
| Credit (t-1) | -0.219 | 0.8721* | 0.5478* | 0.6173* | -0.3097* | -0.2132 | 1.0000 | |||
| GDP(t-1) | -0.0999 | 0.8030* | 0.7714* | 0.5903* | -0.2853* | -0.0495 | 0.6822* | 1.0000 | ||
| CPI (t-1) | -0.2588* | 0.5804* | 0.3297* | 0.9178* | -0.4238* | -0.2151 | 0.6390* | 0.4696* | 1.0000 | |
| Interest rate (t-1) | 0.2508* | -0.146 | -0.0678 | -0.2381* | 0.7256* | 0.1978 | -0.2772* | -0.2042 | -.3581* | 1.0000 |
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
Significance at 10 percent.
Panama: Contemporaneous and Lagged Correlations Across Macro-Financial Variables 1/
(quarterly frequency)
| NPL | Credit | GDP | CPI | Interest rate | NPL (t-1) | Credit (t-1) | GDP (t-1) | CPI (t-1) | Interest rate (t-1) | |
|---|---|---|---|---|---|---|---|---|---|---|
| NPL | 1.0000 | |||||||||
| Credit | -0.2349* | 1.0000 | ||||||||
| GDP | -0.0672 | 0.6887* | 1.0000 | |||||||
| CPI | -0.2326* | 0.6491* | 0.4799* | 1.0000 | ||||||
| Interest rate | 0.1994 | -0.2778* | -0.2064 | -0.3584* | 1.0000 | |||||
| NPL (t-1) | 0.9513* | -0.1897 | 0.0091 | -0.2058 | 0.1401 | 1.0000 | ||||
| Credit (t-1) | -0.219 | 0.8721* | 0.5478* | 0.6173* | -0.3097* | -0.2132 | 1.0000 | |||
| GDP(t-1) | -0.0999 | 0.8030* | 0.7714* | 0.5903* | -0.2853* | -0.0495 | 0.6822* | 1.0000 | ||
| CPI (t-1) | -0.2588* | 0.5804* | 0.3297* | 0.9178* | -0.4238* | -0.2151 | 0.6390* | 0.4696* | 1.0000 | |
| Interest rate (t-1) | 0.2508* | -0.146 | -0.0678 | -0.2381* | 0.7256* | 0.1978 | -0.2772* | -0.2042 | -.3581* | 1.0000 |
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
Significance at 10 percent.
Panama: Variance Decomposition 1/
(quarterly frequency, in percent)
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
Panama: Variance Decomposition 1/
(quarterly frequency, in percent)
| NPL | Credit | GDP | CPI | Interest rate | ||
|---|---|---|---|---|---|---|
| 4-quarter horizon | ||||||
| NPL | 0.9 | 0.0 | 0.1 | 0.0 | 0.0 | |
| Credit | 0.2 | 0.4 | 0.3 | 0.1 | 0.0 | |
| GDP | 0.2 | 0.0 | 0.8 | 0.0 | 0.0 | |
| CPI | 0.1 | 0.0 | 0.4 | 0.5 | 0.1 | |
| Interest rate | 0.1 | 0.1 | 0.1 | 0.1 | 0.7 | |
| 8-quarter horizon | ||||||
| NPL | 0.8 | 0.0 | 0.1 | 0.0 | 0.0 | |
| Credit | 0.2 | 0.3 | 0.4 | 0.1 | 0.1 | |
| GDP | 0.1 | 0.0 | 0.7 | 0.1 | 0.1 | |
| CPI | 0.1 | 0.1 | 0.5 | 0.4 | 0.1 | |
| Interest rate | 0.1 | 0.1 | 0.1 | 0.1 | 0.6 | |
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
Panama: Variance Decomposition 1/
(quarterly frequency, in percent)
| NPL | Credit | GDP | CPI | Interest rate | ||
|---|---|---|---|---|---|---|
| 4-quarter horizon | ||||||
| NPL | 0.9 | 0.0 | 0.1 | 0.0 | 0.0 | |
| Credit | 0.2 | 0.4 | 0.3 | 0.1 | 0.0 | |
| GDP | 0.2 | 0.0 | 0.8 | 0.0 | 0.0 | |
| CPI | 0.1 | 0.0 | 0.4 | 0.5 | 0.1 | |
| Interest rate | 0.1 | 0.1 | 0.1 | 0.1 | 0.7 | |
| 8-quarter horizon | ||||||
| NPL | 0.8 | 0.0 | 0.1 | 0.0 | 0.0 | |
| Credit | 0.2 | 0.3 | 0.4 | 0.1 | 0.1 | |
| GDP | 0.1 | 0.0 | 0.7 | 0.1 | 0.1 | |
| CPI | 0.1 | 0.1 | 0.5 | 0.4 | 0.1 | |
| Interest rate | 0.1 | 0.1 | 0.1 | 0.1 | 0.6 | |
NPL = NPL ratio, Credit = growth in credit to the private sector, GDP=real GDP growth, CPI = total inflation, Interest rate = simple average of commercial interest rate of foreign and Panamanian owned banks.
D. Conclusions
6. Bank asset quality should be closely monitored to prevent the emergence of systemic risks from the build-up of debt in recent years. While NPLs remain low and are well-provisioned, they are likely to come under further pressure as rising global interest rates continue to put pressure on interest rates in Panama’s dollarized economy. Effective supervision will be critical to help prevent a further deterioration of asset quality, while the recent strengthening of provisions following the transition to IFRS9 will help to facilitate recognition of losses.
Annex I. Impact of Credit Growth Shock
Impact of 1 Standard Deviation Increase in Credit Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Impact of 1 Standard Deviation Increase in Credit Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Impact of 1 Standard Deviation Increase in Credit Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Annex II. Impact of Real GDP Growth Shock
Impact of 1 Standard Deviation Increase in Real GDP Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Impact of 1 Standard Deviation Increase in Real GDP Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Impact of 1 Standard Deviation Increase in Real GDP Growth 1/
Citation: IMF Staff Country Reports 2019, 012; 10.5089/9781484394069.002.A005
1/ As estimated from orthogonalized impulse response functions from a VAR model including the NPL ratio, growth in credit to the private sector, real GDP growth, inflation and the commercial paper rate. Source: Author calculations.Prepared by Kimberly Beaton.
The commercial (which includes services) interest rate is calculated as the simple average of the commercial interest rate charged on new lending by foreign and Panamanian-owned banks. Commercial lending represents about 40 percent of new credit.
Unit root tests confirm that all variables are stationary in order I(0).
Ideally, the model would also include the unemployment rate to reflect the fact that high unemployment is likely to lower borrower repayment capacity and lead to a deterioration in asset quality. However, there is no quarterly data on unemployment dynamics available for Panama.
Impulse responses are orthogonalized, using the Cholesky decomposition to identify the orthogonal shocks. The variables are included in the VAR in the following order: the NPL ratio, real GDP growth, inflation, and the commercial interest rate. This ordering reflects our assumption that economic activity affects asset quality with a lag, while asset quality has a contemporaneous effect on economic activity, largely through credit. Estimation results are broadly consistent to the ordering of the variables.
The negative correlation with economic activity and positive correlation with interest rates is not statistically significant.
