It’s not entirely true that banks are run like casinos. But as far as their financial performance was concerned, that seemed to be happening. Their ability to set high prices and the resulting profits proffered other narratives. Indeed during 2010 to 2018, when inflation was also relatively low, banks reported returns (return on equity or ROE) in the range of 16% to 23%.
Casino-like-profits, the argument went, came at the expense of borrowers like small businesses and consumers. It was then commonplace for regulators to argue that banking should be run like a boring utility and not a casino.
New regulations and taxes biffed banks in 2018. Analysts say the impacts of these combined with the shocks from the aftermath of the April 2019’s Easter Sunday attacks will show that most financial institutions have walls of straw.
For banks and finance companies, hiding dodgy loans are now more difficult because of a new accounting rule. They are also now required to have more capital and on top of that even more capital to match higher-risk lending. Since 2018 the government targeted bank profits for higher taxation.
The effects of these, combined with the impact of the economic slowdown, are now visible.
[pullquote]Investors watch ROE and equate the outcome to success or failure. However, ROE is not a good measure for medium-term profitability or long term shareholder returns[/pullquote]
Return on equity had fallen to 13.15% in the December 2018 quarter. ROE was the lowest since 2009 for that particular quarter. Central Bank tabulated data for the March ’19 quarter had not been released at the time of going to print. Some of the largest commercial banks that have released their results for March ’19 have shown that ROEs have declined further.
Net profits of local commercial banks (except People’s Bank and Cargills Bank which were unavailable at the time of writing) in the March quarter had plunged 27% from a year earlier.
Publicly available central bank data indicates this is the steepest commercial bank profit decline in 11 years. The previous record fall was in the March quarter of 2014 when profits fell 21%.
Bankers had anticipated profits would be hit. Some stock analysts had not. Net profits had declined during the previous three consecutive quarters. That fall was easily explained, some thought, by the base effect of IFRS 9 accounting standard adoption which increased loan loss provisions, and new taxes including the derided Debt Repayment Levy and taxes on income from unit trusts and development bonds.
There were other impacts of the new accounting rules too. Under IFRS 9, a bank which owned shares of another had to move those holdings to a trading account if they were held in an investment account previously. When the market price of shares change, losses or gains for any held in the trading account have to be recognised in the financial statement. In case there is a value write-down of these shares, more capital will be required to cover those losses.
Another reason that impacted profitability was the government delaying payments to construction firms. Construction contributes around 7% to GDP. The backlog of payments for government-financed projects increased when for a few months in late 2018 a constitutional crisis sent into limbo all state activity. Contractors, who had obtained loans from banks to fund working capital and equipment, had Rs.90 billion in unpaid bills.
By end-March 2019 much of the construction sector debt had been settled. As a result, banks’ March quarter profits slipping to an 11-year low surprised analysts. Overall these changes should have made the industry safer. Maybe also a bit duller, due to the low profitability.
However, combined with commercial bankings structural problems, low profitability and possible effects of the Easter Sunday bombings, which will only start to be reflected in the June quarter results, may lead to a crisis for Sri Lankan commercial banking.
Bad loans which slowly declined during the last decade are now rising quickly. “It’s not the level of bad loans which is of most concern, but the rate at which it is growing,” Central Bank Governor Indrajit Coomaraswamy said earlier in 2019. Bad debt surged to 3.4% of total loans in the 2018 December quarter compared to 2.5% a year earlier.
The deterioration of loan quality has accelerated since. At Bank of Ceylon (BOC), bad loans grew to 4.66% from 3.62%. In Commercial Bank, it was 4.14%, rising from 3.24%. Hatton National Bank saw the steepest climb, to 4.63% from 2.78%. Sampath Bank’s bad loans rose to 4.87% from 3.69%.
Loan loss provisioning had increased by 78% in the March quarter for the banks which had released March quarter results analysed for this story. The worst was at HNB, with loan losses tripling to Rs.4.6 billion from Rs.1.5 billion. The rock solid image of banks isn’t borne out by the industry credit risk according to an assessment of commercial banking by Standard and Poor’s, a credit rating agency.
Credit rating agency S&P’s Banking Industry Country Risk Assessment for Sri Lanka released end-January 2019, downgraded the industry by one to 9, just a notch away from the worst possible ranking of 10. According to S&P, Sri Lanka’s banking industry is in the same league as those of Kenya, Papua New Guinea, Cambodia, Bangladesh, Vietnam and Mongolia.
The report, which came a few months ahead of the Easter Sunday bombings examines risks in two parts; economic and industry. Sri Lanka’s banks scored 9 on economic risks and 8 on industry risks for the overall rating of 9. The report presents data for Sri Lanka’s largest commercial banks and NSB, which is a government-owned specialised bank. Banking is a business of trust. Depositors and lenders – who fund about 90% of a bank’s business – must trust that they will get their money back. If somehow that trust is broken and everyone wants their money back at the same time a bank would go out of business.
As a result, banks everywhere build an image of stability. As the spike in loan loss provisions proves, sometimes it is just veneer.
The S&P Banking Industry Country Risk Assessment for Sri Lanka is not a credit rating. It’s instead a report that outlines the relative risks of the industry here. It says that the Sri Lankan banking sector’s resilience is weak and is affected by the country’s low-income levels (with an estimated per capita GDP of about US$4,050 in 2018).
“Credit risk is high, in our opinion, given relaxed lending and underwriting standards as well as evolving risk management practices. The credit risk is accentuated by high growth in credit in the past and a recent slowdown in GDP growth,” it said.
There is a simple and liniear link between private credit and economic growth. Private sector credit growth was 14.9% in 2018, and the Central Bank forecast it will expand 13.5% in 2019. So far, private credit grew 11.3% in the March 2019 quarter from a year earlier. Month-on-month credit growth, however, has shown an alarming trend. In January 2019, month-on-month private credit fell, for the first time since May 2014, and has barely grown since. Lower credit demand is a symptom of economic malice and not necessarily the cause. However, regulators make credit cheaper when growth worries take precedence over inflation. The central bank intervened in markets to lower retail lending rates amidst a fall in other benchmark rates. The regulator forced a two per cent lending rate cut to drive credit. Fitch Ratings is expecting small banks to be aggressive in lending. However, analysts predict larger banks will hold off on lending or reduce lending rates, and to later call for the removal on the deposit rate cut.
Chart 1, which plots nominal economic growth against private credit growth, suggests loan demand is rising faster than the economy. However, as banks reduce lending, this is expected to narrow. It is a good thing that new accounting rules and BASEL III capital requirements are forcing big banks to be safer. But the trouble is, that falling profitability will undermine banks’ ability to raise capital. Some banks don’t already meet the capital hurdles. For others, a dip in profitability, rise in dud loans or reclassification of their investment portfolio will take them dangerously to the edge or over it (see Chart 2).
When a bank is undercapitalised restrictions immediately kick in. It is unable to expand its loan book, especially to more profitable areas where lending attracts higher capital risk weights. It will have to come up with a new strategy because it will no longer be able to accept the same lending risks. It will also face the challenge of raising capital at its most vulnerable when its market valuation is low.
Investors watch ROE and equate the outcome to success or failure. However, ROE is not a good measure for medium-term profitability or long term shareholder returns. No bank chief executive can run the institution with an ROE focus, because the measure does not account for risk. Regulations like BASEL III, accounting standards and other rules all force chief executives to focus on managing risks rather than short term returns measured by ROE. As the fallout of the Easter Sunday attacks spread, banks will realise that keeping medium to long term return goals aligned will be impossible if unexpected losses have eroded their capital base.
S&P said measures by the Central Bank to strengthen capital positions in the next few years would benefit the banking system in the longer term by providing suitable capital buffers to absorb unexpected losses. The Easter Sunday attacks happened three months following that statement. Banks have struggled to tap public markets to shore up their capital buffers because of poor valuations to listed equities in Sri Lanka in general. In some months, it will become apparent that few banks had time to build capital buffers to face unexpected losses.
Already, most large banks are investing their funds in gilts, instead of granting loans. The influx of bank funds into the gilt market has seen T-Bill auction yields fall from 9.91% to 8.90% in one month to end-May 2019.
Some banks are also not able to lend due to regulatory obstacles. At the state-owned BOC, which is Sri Lanka’s largest bank with 20% of the industry loan book, Tier 1 risk-weighted capital was at 10.19% by March 2019, dangerously close to the 10% minimum. The government had injected Rs.5 billion capital in 2018, but the bank had paid an equal amount in dividends. BOC’s loan book shrunk 2.1% from December ’18 to March ’19, leading to a contraction of total loans and assets of the largest bank. People’s Bank, also state-owned, tends to follow BOC’s trends, although, in December ’18, People’s Bank’s Tier 1 capital was at a slightly better 11.02%.
The banks are now saddled with loans given to high-risk economic activities. At HNB, 17.8% of loans were to the tourism sector by end-December ’18, while at Sampath Bank it was 9.3%. Construction (hotel constructions are in this category for some banks, and in tourism for others) was 10.3% of HNB loans and 19.4% of Sampath Bank’s portfolio. Commercial Bank was less exposed to sectors impacted by the current economic downturn, with just 6.6% of its loan book in tourism and 4.5% in construction.
While delayed payments on government projects and a downturn in demand for condos are affecting the construction sector, the tourism industry has been under pressure over the past decade. Despite unprecedented growth in tourist arrivals and earnings since 2010, most holidaymakers have been staying at B&Bs and guesthouses and not in big hotels.
Banks have upgraded the risk profile of the construction and tourism industries, and that too is affecting their bottomlines. IFRS 9 was expected to be a one-off hit on the profit and loss account of banks in 2018, and the loan loss provisions were supposed to grow at a natural rate in line with bad loans. However, the standard works in such a way that as risks in specific industries grow and macroeconomic indicators suffer, loan loss provisions, even for good loans, increase.
[pullquote]Banks have upgraded the risk profile of the construction and tourism industries, and that too is affecting their bottomlines[/pullquote]
The banking crisis suggests slow economic growth in the final quarter of 2018 (1.8%) due to the constitutional crisis is likely to have extended into the first quarter of 2019. GDP figures haven’t yet been released. However, the Business Sentiment Index (BSI) of the Central Bank, which surveys industries to reflect their contribution to GDP, showed business condition, profitability, production, demand, and sales falling below expectations.
Expectations have worsened for the second quarter. Demand and sales expectations, which are usually the highest in the second quarter due to the festivities, have fallen to the lowest levels since the index started in 2014.
Analysts forecast the economy will contract in the second quarter of 2019. In April production falls, as most factories close for lengthy new year holidays when people working away from their homes return to their villages.
However, this year most who went home for the holidays’ didn’t return to their city jobs until mid-May, due to the security uncertainty. This has left just one month in the second quarter in which business could normally take place. Some analysts forecast the government will resort to fiscal stimulus during the rest of the year. The Central Bank too is attempting to stimulate the economy with interventions in banks and has cut policy rates in end-May.
The state may also need to recognise that its relationship with the banking sector must change. Recently banks have been treated as hybrid milch cows and whipping boys. At least in terms of their financial performance banks are now looking less like casinos. Their ROEs are expected to be hit by the rising bad debt, low credit demand and their own risk aversion. Their mighty edifices now appear to be made of straw.
[Illustrations by: Akila Weerasinghe]