Brazil’s banks are facing an apparent paradox: the level of corporate leverage has improved markedly in the past year while delinquency levels have risen and non-performing loan (NPL) formation remains at high levels.
These are the main findings of a report into the Brazilian banking sector by Credit Suisse.
Lead analyst Marcelo Telles says the main characteristics of Brazil’s credit crunch peaked a year ago. For example, net debt to last-12-month ebitda improved to 2.2 times for the system – excluding Petrobras debt – in 1Q17 from 3.0 times one year earlier. Interest coverage also improved to 2.7 times, from 1.4 times in 1Q16.
Telles believes the reason behind this apparent contradiction between these main findings lies in the performance of renegotiated and restricted loans than have been agreed in the past couple of years, and that lengthened the NPL cycle and is now translating into higher NPLs.
“Despite improvement in corporate leverage, delinquency in the corporate sector has been deteriorating constantly since early 2015, while that of the retail sector has stabilized many months ago,” argues Telles.
“Since 2015, corporate delinquency increased by 120 basis points, but retail delinquency actually improved by 20bp, despite the economic deterioration since then.
“In our view, one of the reasons for the discrepancy between better corporate leverage and worse delinquency indicators stems mainly from the increase in the volume of restructurings and renegotiations, which ended up creating a backlog of future potential NPLs and making the NPL cycle longer.”
Data from the Brazilian central bank support this interpretation: the size of renegotiated and restructured corporate debt has more than doubled since 2015 to 2016, reaching 9.9% and 1.9% of total loans – from 4.2% and 0.9% at the beginning of 2015 – respectively.
However, the banks have been reacting well to the deterioration in asset quality and increased provisions to improve the coverage of these ‘troubled’ loans. Coverage now stands at 21.8% for the system (up from 13.2% in 1Q2016) – with private banks having a 35% coverage ratio – which is approaching cyclical norms; in line with 2011’s coverage ratio and close to the 10-year average of 25.8%.
This suggests there could be an improvement in the cost of risk in the coming quarters, albeit gradually given the still-high leverage ratios. Credit Suisse expects that “the cost of risk should normalize gradually and probably returning to 2013 and 2014 [levels] by 2019”.
Increased corporate leverage in Brazil in the couple of years up to 1Q16 led to a significant increase in the cost of risk of commercial loans during that period. The main component of this increased cost of risk was the need for increased provisioning by the Brazilian banks since the end of 2014.
In 2016, provisions for commercial loans comprised 44.5% of total provisions for Itaú – compared with 31.3% in 2014. For Banco do Brasil, provisions for commercial loans jumped from 58.8% in 2014 to 74.7% in 2016.
The wide discrepancy in the size of provisions needed to cover troubled loans to the corporate sector means that the recent improvement will also lead to uneven improvements among Brazil’s largest banks.
Credit Suisse believes that Bradesco and Banco do Brasil have the most room to improve the cost of risk, while better credit control through the cycle from Itaú and Santander means that those banks have less room to enjoy significant reductions in their cost of risk.
Expectations of further cuts to Brazil’s base rate – the Selic rate has been cut from 14.25% to 10.25% between October and May – should leave to further improvements in corporate leverage in the second half of this year, which is when the easing cycle is expected to come to an end.
Further progress in deleveraging after this point will require growth in ebitda, which in turn relies broadly on GDP growth, and the outlook for that keeps being reduced.
The consensus for 2017 is now under 0.5% growth and with all the political uncertainty and questions about the current government’s reform agenda, this is a clear unknown.
“Ebidta growth is key for the deleveraging of the corporate sector and largely dependent on GDP,” according to Telles, who incorporated GDP growth of 2.0% into his model’s base case.
“If GDP growth does not recover within the next two years, net debt-to-ebitda ratio would remain high at 1.92 times in 2019. For 0.5% less of GDP growth, net debt-to-ebitda rises by 4.2%. The impact of GDP growth is higher than monetary easing as for every 50bp decline in the Selic rate, leverage declines by 1.3%”.
Of course, the sensitivity analysis of the two components cannot be considered in isolation as they will combine to enhance the positive or negative effects – as lower rates will prompt higher GDP and vice versa.