articles Corporate /en/research-insights/articles/2016-annual-sovereign-default-study-and-rating-transitions content
BY CONTINUING TO USE THIS SITE, YOU ARE AGREEING TO OUR USE OF COOKIES. REVIEW OUR
PRIVACY & COOKIE NOTICE
Log in to other products

Login to Market Intelligence Platform

 /


Looking for more?

Request a Demo

You're one step closer to unlocking our suite of comprehensive and robust tools.

Fill out the form so we can connect you to the right person.

  • First Name*
  • Last Name*
  • Business Email *
  • Phone *
  • Company Name *
  • City *

* Required

In This List

2016 Annual Sovereign Default Study and Rating Transitions

S&P Global Platts

Energy: What to Watch in 2019

S&P Global Ratings

COP24 Special Edition Shining A Light On Climate Finance

S&P Dow Jones Indices

Considering the Risk from Future Carbon Prices

Banks In Emerging Markets 2019 - The Mid East, North Africa, & Turkey


2016 Annual Sovereign Default Study and Rating Transitions

In 2016, two defaults occurred among the sovereign obligors that S&P Global Ratings rates. This was one more default than in 2015 and the fifth consecutive year with at least one sovereign default. Including the default, we lowered 24 sovereign ratings and raised nine in 2016. Rating actions were less numerous than in 2015, but the number of downgrades was the highest since 2011.

Overview

  • The rank ordering of sovereign ratings has been consistent with historical default experience, and these ratings continue to serve as effective indicators of relative credit risk.
  • Of the 130 active sovereign ratings at the end of 2016, 52% were investment grade.
  • Sovereign ratings generally exhibited greater stability at higher rating levels than at lower levels, as we would expect.
  • The ratio of downgrades to upgrades increased to 2.67x in 2016 from 1.40x in 2015. In addition, the average number of notches of foreign-currency downgrades increased to 1.50 from 1.15. The average number of notches for upgrades remained the same at 1.00.

We track rating movement according to the number of ratings that changed during 2016 as opposed to the number of times a rating changed. For example, this study considers a rating that was lowered twice in 2016 to reflect one lower rating at the end of 2016 than at the beginning of 2016. On the other hand, if a rating was lowered and later raised to the rating at the start of the year, no rating action would appear in this study.

This study is based on long-term sovereign credit ratings. The methodology tracks rating migrations over time and includes revisions to 'SD' (selective default). An 'SD' rating is more common for sovereign issuers than a 'D' rating because defaulting sovereigns often continue to service some of their debt. This is an issuer ratings-based study, as opposed to being based on issue ratings. In other words, we look at the sovereign ratings on the central governments themselves, not the ratings on the individual securities these governments might have issued.

Our metrics treat all issuers equally and are not adjusted for size or influence. Therefore, for the purposes of this study, a default by Argentina counts the same as a default by Mali, even though the latter has a much smaller economy. Our study tracks defaults on a sovereign's commercial debt, including both bonds and bank loans.

Withdrawn ratings (as indicated with the abbreviation 'NR,' which stands for "not rated") are included up until the date of withdrawal. We record defaults after the date of withdrawal if we obtain knowledge of those defaults. As of Dec. 31, 2016, S&P Global Ratings had withdrawn 11 public sovereign ratings: Benin, Cambodia, Gabon, Guernsey, Isle of Man, Kyrgyz Republic, Libya, Madagascar, Mali, Seychelles, and Tunisia. (We reinstated the rating on Guernsey in October 2014.) There are a total of 166 rating records for sovereigns, including those of defaulted ratings.

The number of sovereigns that we rate has grown as more governments access the international bond markets. We have rated 141 sovereigns (foreign currency) since 1975, and 130 of these had active ratings at the end of 2016.

In the 1990s, speculative-grade-rated sovereigns became more common as we started rating smaller and less-diversified economies. The percentage of speculative-grade ratings significantly increased to 37% in 1999 from 3% in 1991. This was mostly a result of new ratings being assigned, as the number of speculative-grade ratings rose to 30 from two, while the number of investment-grade ratings increased to 52 from 29. During 2000-2009, speculative-grade ratings continued to outpace investment-grade ratings, but the latter group grew as well. Speculative-grade ratings increased to 53 in 2009 from 33 in 2000, while investment-grade ratings grew to 70 in 2009 from 54 at the start of the decade. The percentage of speculative-grade ratings was 43% in 2009. During 2010-2016, the number and percentage of speculative-grade ratings increased to 62 and 48%, respectively, while the same for investment-grade ratings declined to 68 and 52%. In 2016, three investment-grade sovereigns were downgraded to speculative grade, while one speculative-grade sovereign was upgraded to investment grade.

Seventy-one of the initial sovereign ratings we have assigned have been speculative grade; we assigned the first to Hungary in April 1992. At the end of 2016, 11 of the 71 sovereigns with initial speculative-grade ratings were rated investment grade, 18 defaulted at some point, and nine had withdrawn ratings. On the other hand, of the 69 sovereigns initially assigned investment-grade ratings since 1975, seven were rated speculative grade at the end of 2016, four defaulted, and three were withdrawn. Of the 130 sovereigns with active foreign-currency ratings at the end of 2016, 68 (52%) were investment grade and 62 (48%) were speculative grade. As of March 17, 2017, three sovereign ratings were rated one notch above speculative grade ('BBB-') with negative outlooks: Kazakhstan, Oman, and South Africa. Among the speculative-grade sovereigns, one was rated one notch below investment grade ('BB+') with a positive outlook: Indonesia and Russia.



Energy: What to Watch in 2019

Highlights

S&P Global Platts Analytics Issues Two Special Reports

Pricing across the global energy markets will face headwinds in 2019, with a weaker and more uncertain macroeconomic framework deflating price formation in general, according to two special reports just issued by S&P Global Platts Analytics. Such headwinds will require the industry and portfolio managers to take a big-picture approach.

See the Executive Summary of the S&P Global Platts Analytics special report 2018 Review and 2019 here. Access the full S&P Global Platts Analytics Top Factors to Look Out For in 2019 for Energy here.

"One of the key lessons learned in 2018, painfully by some, is that market sentiment can shift violently without much change in fundamentals, requiring a steady, holistic perspective," said Chris Midgley, global head of analytics, S&P Global Platts. "It is clear that this volatility will remain a feature across the energy markets in 2019, particularly as IMO 2020 nears."

Particularly blustery headwinds are in store for markets where prices finished 2018 at elevated levels, and well above costs, such as North American natural gas and global coal. However, if the supply side can adjust to the reality of slowing demand growth, energy prices can find support. For natural gas liquids (NGLs), the ongoing logistical constraints at the US Gulf Coast are likely to manifest on continued price volatility, particularly for ethane and liquid petroleum gas (LPG), over the next year despite strong global demand.

LPG, such as propane and butane and used in transportation fuel, refrigeration, heating and cooking, is rapidly facing US export capacity constraints, especially along the US Gulf Coast. For LPG feedstock propylene, there is clear potential for high volatility globally over the next 12-18 months.

Analysts at S&P Global Platts see weakening prices of Henry Hub natural gas. The slowdown in US demand growth will exceed that of supply. But if winter temperatures prove to be colder than normal, near-term prices will need to move higher to bring on enough supply to replenish depleted storage levels.

For global liquefied natural gas (LNG), it will be end-user-backed LNG demand that faces particular struggle to cope with the speed and force of new supply entering the market in 2019. Non price-responsive demand in Asia will be easily met and JKM spot physical prices (reflecting LNG as delivered into Japan, Korea and China) will sag next year.

Access the full S&P Global Platts Analytics Top Factors to Look Out For in 2019 for Energy here. Among the 22 key take-away themes:

  • NGL supply growth will strain the North American energy system
  • Saudi Arabia will need to be nimble to balance 2019 oil supply
  • US oil supply limited by pipelines
  • Oil demand slowing: trade war, industrial slump
  • 2019 LNG supply additions largest since the Qatari mega-trains
  • US gas supply growth to exceed demand growth even with LNG exports
  • Global solar growth slowing
  • Shipping disruption looming - IMO 2020
  • New Russian gas pipeline advantage over Ukraine
  • US coal demand to decline again in 2019
  • Growth in new refineries and complex capacity likely to weigh on refinery margins especially in Asia

Year 2019 will certainly be one of transition for crude and refined oil products as it will lead into 2020 when roughly three million barrels per day of high-sulfur fuel oil must be “destroyed” (including enhanced usage of HSFO in power generation) due to the International Marine Organization (IMO) mandate of eco-friendly shipping fuels in use at sea. A similar amount of middle distillate/low sulfur fuel must be created (by refinery changes and by running more crude oil. The increase in refinery capacity between now and 2020 is large, but mostly needed to cover normal demand growth. Expect prices of light sweet crudes to be bid up in 4Q19.



COP24 Special Edition Shining A Light On Climate Finance

Highlights

− Green loans are evolving, with the Climate Bond Initiative forecasting nearly $1 trillion in green bond issuance by 2020.

− Despite the uptick in green bond and loan issuance, the market still remains relatively small, especially compared to the universe of assets comprising CLO 2.0 transactions.

− In our view, a green CLO market has large growth potential, boosted by regulatory initiatives and emerging interest from both issuers and investors in 2018.

− We built a hypothetical rating scenario for a green CLO to compare and contrast the underlying portfolio and structure with a typical European CLO 2.0 transaction.

− Our hypothetical green CLO analysis showed that green loans may have different fundamental characteristics to corporate loans, such as lower asset yields, higher credit quality, and higher recovery rates assumptions.

The global collateralized loan obligation (CLO) market has experienced a rebirth (2010 in the U.S. and 2013 in Europe). New issuance continues to increase due to investor familiarity with the product, as well as low historical default rates. While a market for green assets, such as green loans and bonds has been established for a while, although still of a relative size, a sustainable securitization market is still in its infancy. Considering the challenge in financing the amounts, S&P Global Ratings expects green CLOs to play a role in increasing the private sector presence in the sustainable finance market.

Following the Paris Agreement that came into force in November 2016, 184 parties have ratified the action plan to limit global warming. For this purpose, developed nations have pledged to provide $100 billion (about €87 billion) annually until 2025. As part of this deal the EU has committed to decrease carbon emissions by 40% by 2030. In March 2018 the European Commission (EC) proposed the creation of environmental, social, and corporate governance 'taxonomy', regulating sustainable finance product disclosures, as well as introducing the 'green supporting factor' in the EU prudential rules for banks and insurance companies.

Read the Full Report
Download


Considering the Risk from Future Carbon Prices

Along with the advent of the 2015 Paris Climate Agreement has come a growing understanding of the structural changes required across the global economy to shift to low- (or zero-) carbon, sustainable business practices.

The increasing regulation of carbon emissions through taxes, emissions trading schemes, and fossil fuel extraction fees is expected to feature prominently in global efforts to address climate change. Carbon prices are already implemented in 40 countries and 20 cities and regions. Average carbon prices could increase more than sevenfold to USD 120 per metric ton by 2030, as regulations aim to limit the average global temperature increase to 2 degrees Celsius, in accordance with the Paris Agreement.

S&P Dow Jones Indices launched the S&P Carbon Price Risk Adjusted Indices to embed future carbon price risk into today’s index constituents.

Read the Full Report
Download


Banks In Emerging Markets 2019 - The Mid East, North Africa, & Turkey

Highlights

- Difficult economic and political conditions will likely continue to test MENAT banks this year.

- While we anticipate loan growth to average 7%-8% in 2019, we don't think this level, if adjusted for inflation, is sufficient to finance the region's development needs.

- Weak asset quality will continue to weigh on our view of the credit quality of banks in the region, particularly those in Turkey.

- High dependence on foreign funding remains one of the most prominent risks. While some banks still rely on nonresident transfers, which we expect to continue growing, others are more dependent on volatile wholesale external funding.

Jan. 17 2019 — After a difficult 2018, banks in the Middle East, North Africa, and Turkey face more of the same this year due to tighter global liquidity conditions, a stronger U.S. dollar, and geopolitical as well as local instability. S&P Global Ratings believes the region's prospects for economic growth will be dampened, notably by Turkey's expected contraction (-0.5% real GDP growth in 2019). While a weaker Turkish lira has put intense pressures on those private-sector borrowers indebted in foreign currency debt, it has also benefited Turkey's export sector, including tourism, which posted very strong results last year. Nevertheless, much of the lira's depreciation has passed through into higher inflation. A 26% minimum wage hike scheduled for this year is likely to push prices up further, cutting into consumers' already weakened purchasing power. On a positive note, lower commodity prices could provide some breathing space, as most of these countries are commodities importers. We expect oil prices to stabilize around $55 in 2019-2020.

Chart 1  

We expect loan growth to stabilize around a nominal 7%-8% on average in 2019, ranging from 0% for Turkey to 17% for Egypt. However, we consider these figures, if adjusted for inflation, as insufficient to cope with the economic development needs in the Middle East, North Africa, and Turkey (MENAT). Several economies in the region, in particular Jordan and Lebanon, are facing an overall stagnation in lending activities. The Syrian conflict in particular, coupled with social domestic tensions, generally depressed tourism and trade activities, notably for Lebanon and Jordan, which are finding it difficult to process large inflows of refugees that dwarf those into Europe and the U.S. Upcoming elections in Turkey and Tunisia will be critical for the stability of their economies and banking sectors, as they may motivate governments to actively interfere in monetary and other policy settings. Some scenarios could set off another round of market volatility, including exchange rate volatility in Turkey. That could create further difficulties for corporate loan quality.

Chart 2  

Asset quality will continue to weigh on our view of the credit quality of banks in the region. This is particularly relevant for Turkey, where we estimate that problematic loans (Stage 2 and Stage 3 loans) could climb to about 20% of total loans. Tunisia is also another country where we continue to view asset quality negatively. While banks' reported nonperforming loans (NPLs) dropped slightly in the past two years, we think asset quality indicators could be worse if banks were to adopt International Financial Reporting Standards 9 (IFRS 9).

High dependence on foreign funding remains one of the most prominent risks for MENAT banks, in our view. While some still rely on nonresident transfers (Morocco, Lebanon, Jordan, and Egypt), which we expect to continue growing, others are more dependent on volatile wholesale external funding (Turkey). On Sept. 30, 2018, the total external debt of Turkish banks stood at $100.4 billion ($117.8 billion including nonbank financial institutions) including $40.5 billion coming to maturity in the next 12 months ($47.0 billion including nonbank financial institutions). These numbers exclude other short-term liabilities, including, among others, nonresident deposits (both in Turkish lira and foreign currencies), which stood at $60.6 billion on Sept. 30, 2018. The volatility of the Turkish lira and uncertainty related to policy direction, as well as the deterioration of the country's relationship with Western allies, weighed on investor sentiment toward Turkey and led to the default of some corporates and nonbank financial institutions. Market access has somewhat improved recently, with several banks refinancing their debt with high rollover rates, albeit at higher costs. It is important to note that banks still have some foreign currency-denominated assets (about $95 billion on Sept. 30, 2018). If rollover rates drop below a certain level or banks lose a significant amount of other short-term debt, the risks will be displaced to the balance sheet of the central bank because banks will have to use their foreign currency reserves deposited there.

Chart 5  

We expect return on assets to decline slightly for MENAT banks in 2019, to an average of about 1.2%. Some banks have benefited from higher government bonds yields (particularly in Egypt, Lebanon, Tunisia, and Turkey). Others have continued to benefit from low cost-to-income ratios in a global comparison, given the low cost of labor in some systems. We expect both trends will continue in 2019. However, we believe cost of risk will increase slightly in some systems in 2019, such as Turkey and Tunisia. In Turkey, we think the backlog of problematic assets will contribute to higher cost of risk, while in Tunisia the decline in real estate prices will push some banks to beef up their provisions against exposures of real estate developers.

Chart 8 

The development and sophistication of regulatory frameworks for banks remain disparate across the region. Tunisia and Egypt lag behind the rest of the region that has moved toward a stricter regulatory environment under Basel III, sometimes with specific adjustments (Morocco for example). Three out of six countries started implementing IFRS 9. Only Morocco is pursuing the implementation of a recovery and resolution regime. We assess the banking authorities in Tunisia as the weakest among MENAT countries. Although several reforms have been implemented in the past five years, banks continue to report under Basel I and local accounting standards, for example.

Egypt: High exposure to the sovereign and low financial inclusion mean significant untapped potential

In 2018, S&P Global Ratings upgraded the sovereign rating on Egypt to 'B' from 'B-' and revised upward the Banking Industry Country Risk Assessment (BICRA) to group 9 from 10 (on a scale of 1-10 with 1 being the lowest risk), on the basis of low loan leverage, strong liquidity in the domestic banking sector, and supportive Central Bank of Egypt policies through injections of Tier 2 capital. This led to higher ratings for three rated domestic banks. We expect Egypt to experience strong economic growth averaging over 5% from 2019 to 2021, with a smaller current account deficit relative to 2015-2017 and rising investment. Egypt's external position is improving on the back of rising domestic gas production, which curbs imports, and a rebound in tourism. While external financing needs remain high owing to higher levels of short-term external debt, we expect them to decline gradually over the next three years. However, we expect fiscal challenges to remain considerable but on a downward trajectory. High interest rates and higher currency depreciation or oil prices than we expect could derail government efforts to reduce gross government debt from around 90% of GDP currently.

We expect loan growth in 2019 to stabilize at around 15%-18%, compared with around 50% in 2017, primarily because of the devaluation of the Egyptian pound and government-directed lending. We also think that NPL ratios will stabilize around 5% (or about 8%-9% including restructured loans or asset swaps). Nevertheless, given the inflationary pressures and their negative impact on borrowers' debt-servicing capacity, we expect credit losses to increase to 150 basis points (bps) in the next 12-24 months, from 135 bps in 2017. We anticipate the NPL coverage ratio to remain around 100% over the next 12-18 months.

The loan-to-deposit ratio remains low in Egypt (about 40% as of October 2018). Financial inclusion is low: the stock of banks' credit to the domestic private sector stood at less than 30% of GDP at end-June 2018. Excess funds are channeled to domestic government debt, creating a direct link between government's creditworthiness and banks'.

Jordan: The banking system remains vulnerable to regional geopolitical risks

We expect Jordan's GDP growth will reach 2.8% by 2020, up from 2% in 2018 but much lower than the 6.5% rate from 2000 to 2009. Political developments in Syria and Iraq have led to a high financial burden related to the significant influx of refugees and to the closing of former trade routes. This has weighed on economic dynamism and translated into high unemployment and sluggish growth. Over the past couple of years, the reopening of the Syrian-Jordanian and Iraqi-Jordanian borders has led to a slight revival of trade activities. Therefore, we expect a slight rebound over the next four years, permitted by a rise in exports and investment projects, notably in the energy and construction sectors.

The current macroeconomic and geopolitical situation has taken its toll on the banking system as well. With the exception of the sizable cross-border operations of Arab Bank PLC, Jordan's largest bank, the domestic banking system focuses on pure domestic commercial banking operations, both conventional and Islamic. The low-growth environment, higher interest rates, and rebalancing of lending to the private sector from the public sector will result in deteriorating credit conditions for Jordanian banks. Therefore, we expect increased NPLs, though stable profitability thanks to higher margins. However, considering the Jordanian banking sector's track record and regulatory framework, we do not expect a large increase in the cost of risk over the next two years, but rather a slight pickup to about 100-120 bps.

On a positive note, we believe banks' profitability will be sufficient to absorb the rising costs of risk. This is based on an ample customer deposit base, enabling a stable and less costly funding source. We also expect steady moderate lending growth rates encouraged by the recent pro-growth policy led by the central bank.

Finally, we continue to see the banking sectors' significant exposure to sovereign debt (around 20% of total assets) as additional credit risks.

Lebanon: Geopolitical uncertainties and exposure to sovereign debt are the main risks

In November 2018, we revised downward our BICRA on Lebanon to group 10 from group 9, on the basis of significant market distortions created by the Banque du Liban's (BDL) stimulus packages and swaps operations. Although this weakened the stand-alone credit profile of two of the Lebanese banks we rate, all issuer credit ratings remained unchanged at the level of our 'B-' sovereign rating. We project the Lebanese economy to grow by about 2.5% over 2018-2021, increasing from 1.4% in 2017, but lagging behind the levels seen in 2007-2010 (9.2%). However, we believe that large fiscal deficits and increasing public debt will continue to characterize the Lebanese economy in the next two years. Net government debt represents 124% of GDP, which we expect to increase further to 156% by 2021. Given rising global interest rates and external financing costs as well as uncertainty about policymakers' ability to address macroeconomic reforms, we continue to see the government debt as burdensome for the banking sector both directly and indirectly via placements with the central bank offered as part of its "financial engineering" deals. Lebanese banks have high direct exposure to sovereign credit risk. That said, we do not expect a deterioration in domestic economic conditions, excluding unexpected major political instability.

The central bank started reducing its involvement in the economy in 2018. The reduction in BDL's economic stimulus package and subsidies will lower growth in the short term, mostly affecting the real estate and construction sectors. We believe the current operating environment, coupled with an increase in interest rates, will lead to a contraction of lending portfolios and higher deterioration of asset quality, from NPL ratios of 5.7% as of year-end 2017. Nevertheless, we expect the banking sector's profitability to remain higher and more stable than for other sectors in the Lebanese economy. This is mostly based on BDL's support by offering various instruments with effective returns in excess of those on the sovereign debt, but also due to Lebanese banks' significant geographical footprint. The Lebanese banking sector's return on equity has been about 11%-12% over the past few years, a level we expect to stay stable for the next couple of years.

System funding has so far been supported by a highly solid deposit base, with a pronounced nonresident component as well as a high foreign currency-denominated share (69.5% on October 2018). Deposit inflows have proven to be resilient in the past, as most are coming from the Lebanese diaspora.

Political risk remains one of the highest concerns for Lebanon. We forecast that political deadlock will continue to obstruct policymaking. Lebanon held parliamentary elections earlier in 2018 after almost a decade of political turmoil, but the process of forming a government was still on hold as of mid-January 2019, delaying the implementation of structural reforms and preventing the country from tapping pledged donor funds. We expect sectarian divisions and rising tensions among neighboring countries to further weaken the political landscape and weigh on economic growth.

Morocco: Stable banking industry, but pressure on asset quality persists

We expect real GDP to decelerate to about 3.2% in 2019 before accelerating to around 4.1% over 2020-2021. The expanding automotive and tourism sectors, coupled with increased demand for phosphates and derivatives, will stimulate the long-term performance and somewhat reduce the reliance on cyclical sectors, notably agriculture. However, we consider that high unemployment levels, low revenues, and high-income disparity across the country will continue to fuel social tensions. As such, we view positively the government's willingness to make the development of the economy more inclusive.

The Moroccan banking sector remains constrained by its high exposure to commercial real estate risks. This is reflected by overleveraged developers who face decreasing real estate transactions. We forecast overall NPLs to remain high over the next two years (around 7%), with steel, tourism, commercial real estate, and construction bearing the highest credit losses. Moreover, we do not exclude that the ongoing effort by the Central Bank of Morocco (BAM) to harmonize its loan classification and the increase in the general provisions requirement could lead to a decline in asset quality indicators for Moroccan banks.

Profitability is set to remain high in the next two years, despite the increasing provisioning required by IFRS 9 (adopted in 2018), owing to low costs of labor and to a large share of noninterest-bearing deposits (67% of the deposit base as of mid-2018). We forecast that credit growth will be steady at 4% over 2019-2020, based on the recovery of corporate growth and still-dynamic household lending. Though margins will suffer from low interest rates and tough competition in Morocco's banking sector, we anticipate that African subsidiaries will continue to contribute positively to the financial performance of larger Moroccan banks.

We expect customer deposits inflows, especially from Moroccans expatriates, to continue to fully fund credit growth, given banks' limited access to external funding. We expect the latter to remain marginal, notably due to the high costs of fundraising for Morocco in the international capital markets. We believe banks will continue to use domestic capital markets to match increasing funding needs and reinforce their capital base.

In early 2018, BAM widened the fluctuation bands between the dirham and the currency basket to +/-2.5% (versus +/-0.3%). We expect the transition to be long and gradual (15 years) and have limited impact on banks as most financial institutions have marginal exposure to foreign currency-denominated assets.

Tunisia: Subdued economic conditions and overreliance on central bank liquidity remain the main challenges for Tunisian banks

Over 2018, the Tunisian economy showed some slight signs of economic pickup on the back of better agricultural output and a strong rebound in tourism. However, this has not been sufficient to counterbalance the country's high current account deficit, unemployment ratios, and government debt. In our base case scenario, we forecast current economic conditions to persist over the next 12-18 months, with GDP growth averaging 2.8% annually and a gradually narrowing in fiscal and current account deficits that will remain significant. This is based on more aggressive reform implementation, which we expect to continue under the existing government. That said, with the upcoming 2019 election, political uncertainty is increasing. The risk is exacerbated by the chronical dependence of the country on donor financing and the link between the availability of this funding and reform implementation speed.

As for asset quality, Tunisian banks have one of the highest NPL ratios in the region, which we expect to climb to 18% over 2019-2021, incorporating restructured and other tourism sector loans not classified as nonperforming and anticipating increasing financial stress for borrowers as interest rates move higher.

Private-sector leverage remains high. We forecast it will remain at about 85% of GDP in 2019 due to hawkish loan growth over the next three years. Even though Tunisian banks don't use complex products and exhibit a somewhat moderate risk appetite, net margins will remain constrained by price competition.

Although Tunisian banks are mainly funded by customer deposits, the mismatch between their assets and liabilities persists, as the pace of deposit growth is not sufficient to fund the system's lending portfolio. We do not expect to see a reversal of the trend--at least in the short term, mainly because of a growing shadow economy. As such, we expect banks to continue relying on central bank refinancing. We also believe the new rule for the loan-to-deposits ratio as well as haircuts on eligible assets, coupled with expected moderate loan growth, will help to gradually reduce the volume of refinancing.

Turkey: Asset quality and profitability will remain under stress in 2019

Last year was extremely challenging for the Turkish banking system. In August 2018, we lowered our foreign currency sovereign rating on Turkey to 'B+' from 'BB-', on the back of steep depreciation in the Turkish lira, which hurt fiscal balances and financial asset quality. At the same time, we revised lower the BICRA on the country to group 9 from group 7, on our view of increased credit risk in the economy as well as weakening institutional framework and industry stability. As a result, we downgraded the long-term ratings on six Turkish banks.

Political concerns and geopolitical turmoil over the past two years have tested the Turkish economy, leading to a significant decline in the lira over August 2018. Inflation in Turkey rose above 25% in October 2018, and we expect the figure to only gradually decline to 9% over 2020. The central bank fought inflation by raising the rate on the late liquidity window by 11 percentage points to 24% during 2018. However, we believe the volatility of the Turkish lira may undermine the central bank's efforts to anchor inflation, particularly if domestic or geopolitical instability flares up in coming months. This is especially true given the concerns about the recent degradation of institutional checks and balances, including the independence of the central bank: President Erdogan's centralization of power around the central bank brings into question the overall independence of the banking sector and the country's rule of law. The government's response to the recent financial market volatility has been modest, potentially undermined by upcoming elections in March 2019, which limited the rise in interest rates in recent months.

Over the next two years, we expect a reduction in Turkish banking system business volume, and higher credit and funding costs, in line with a contraction of the economy of -0.5% in 2019, before a return to modest growth of 3% by 2020. Banks have proven to be resilient so far, but this resilience hinges greatly on external funding and investors' confidence. We forecast loan growth of 0% in 2019, and we expect that the higher funding and credit costs--due to interest rate spikes, partly compensated by inflation-linked bonds--will weigh on bank profitability over the next 18-24 months. The asset quality deterioration will also hurt banks' profitability, as lenders will need to increase provisions. In addition to the impact of economic contraction and private-sector deleveraging on asset quality, currency depreciation limits borrowers' repayment and debt-servicing capacity given the high share of foreign currency corporate loans (about 40% as of June 2018). Hence, we expect to see reported NPLs to double over 2019 and 2020 to about 6% of total loans. Overall, we estimate that problematic loans (Stage 2 and Stage 3 loans) could climb to about 20% of total loans. We consider the energy, commercial real estate, and construction sectors will continue to represent credit risks for the banking sector.

The capitalization of Turkish banks is also weakening due to lira depreciation (since capital is largely lira-denominated while risk-weighted assets have a large foreign exchange portion) and mark-to-market losses on government securities. Regulatory forbearance initiatives aim to alleviate such pressures. In August 2018, the regulator introduced measures that eliminated mark-to-market losses from lira depreciation and higher rates on capital. Banks are trying to issue hybrid instruments, but there is constrained investor appetite.