Galim Khussainov
Inflation and the tengeBankingLending and mortgagesTaxationEconomy and public policyManagement

Lending and mortgages · 5 min read

Why Business Loans Are Expensive: The Cost of Risk and Debt Recovery

Translated from Russian with the help of AI. Original: Russian version

Business loans are expensive in large part because of the cost of risk built into interest rates. The state can reduce it by speeding up recovery of problem debt, giving banks access to government data on borrowers, and tightening requirements for auditors, appraisers and accounting.

I am often asked why interest rates for business are high, loan maturities are short and banks are reluctant to lend to business, while consumer lending is growing rapidly.

This is a very complex question to cover in a single post, so it is better answered in several. Today I will write about loan pricing and how the interest rate can be lowered by reducing the cost of risk.

The interest rate for any borrower is determined as follows:

The cost of risk-free placement of money (usually proxied by yields on government securities), or the bank's cost of funding if that is higher than government securities yields, + the bank's margin + origination costs + the cost of risk.

The cost of funding depends on inflation, which I will address next time.

The bank's margin depends on market competition, and origination costs depend on automation.

Today we will talk about the cost of risk.

Here is how banks work in lending. They raise money from depositors, various funds, the state, and those who hold banks' bonds and shares.

They then lend this money to households and businesses. In essence, banks are intermediaries. A reasonable question arises: why can't people lend directly to those who need money? The answer is simple: households and businesses need a guarantee that their money will be returned, so banks, as intermediaries, build mechanisms for assessing risk. The higher the risk, the greater the probability of non-repayment.

To cover risks, the volume of expected unrecovered amounts is built into the cost of risk as a percentage added to the loan rate. In other words, all defaults are in effect paid for by the borrowers who do pay. Hence, the higher the risk, the higher the interest rate.

For example, suppose loans of KZT 1 million each are issued to 1,000 people for one year, and 10 of them fail to repay. The 1% in losses must then be covered by the other borrowers. So 1 percentage point is added to the overall interest rate.

But how do you determine how many people will not repay? Banks use data from past years and determine how various parameters affect repayment: for example, credit history, income, collateral coverage, region and so on. These models are then extrapolated into the future. The more requirements are imposed on a borrower, the lower the risk.

Conversely, the fewer the requirements, the more borrowers get loans, but the higher the cost of risk.

So to reduce the cost of risk as a component of the interest rate, there must be fewer defaults and more recoveries from problem borrowers at the same lending volumes.

This is determined by the quality and speed of borrower analysis, and by the quality and speed of debt recovery.

All of these parameters depend on external and internal factors.

Internal factors depend solely on the quality of a bank's risk management, whereas external factors depend entirely on the state.

The state's task, therefore, is to reduce these factors that drive the cost of risk in the banking sector, and the interest rate will fall with them.

Let us look at where the state can influence these parameters.

The first and most important parameter is the speed and quality of recovery of problem debt.

Take an example: out of KZT 10 billion lent to companies, KZT 1 billion is problem loans that are 100% secured by collateral at accepted valuations.

That KZT 1 billion is a non-performing loan. Currently, the average time to recover a problem loan from a company ranges from 1 to 5 years, and in some egregious cases 10–15 years. The average is 3 years.

At a market rate of 15% per annum, the bank loses KZT 450 million on that KZT 1 billion, or 4.5% of the entire portfolio. These losses will therefore be added to the cost of risk.

Recovery depends on court rulings, the work of private bailiffs, the auction procedure, the procedure for taking assets onto the bank's balance sheet, bankruptcy procedures and so on. In my own observation, the recovery period can easily be cut to 1.5 years by putting the judicial system in order. This requires uniform judicial practice, the indisputability of the debt, and no possibility of appealing private bailiffs' actions during recovery. The bankruptcy law must be changed and the bankruptcy procedure simplified, as must the procedure for banks to take collateral onto their balance sheets.

In this way, the volume of non-performing loans can be reduced significantly, since faster recovery procedures will clear problem loans off the books.

Speed also affects the quality of collateral. Without going deeper into the quality of valuations, any property ages and depreciates. Even real estate loses roughly 20–30% of its value over three years of recovery proceedings, which in turn feeds into the cost of the loan.

Thus, to reduce the cost of risk, and with it the interest rate on loans, recovery times must be shortened by improving the work of the courts, private bailiffs and so on.

The next problem is the quality of borrower analysis.

Many people ask why consumer lending is growing today. The answer is simple: there is data for building high-quality scoring models (credit histories, the State Pension Payment Center database, data on individuals from eGov and so on). Because loans to individuals are relatively small, court rulings are issued quickly and consistently, and judges and private bailiffs are not biased. The result is high-quality analysis and a predictable cost of risk, which is built into the interest rate.

For companies, there is nothing of the kind. The quality of information is extremely poor, and verification takes a great deal of time and money. Even companies that undergo audits, unfortunately, fall far short of standards of transparency and openness.

Here, work must proceed in two directions.

The first is to give banks greater access to information on borrowers from government databases, in particular data from the State Revenue Committee, the customs authorities, the justice authorities and so on.

The second is to raise requirements for audit firms and appraisal firms, as well as for bookkeeping and accounting.

The opacity of tax and accounting records stems from entrepreneurs' desire to optimize tax payments and from the need to cash out funds for certain purposes. There is a vast amount of work to be done in this area.

To sum up, reducing interest rates requires reducing the cost of risk, and that depends not only on the bank but also on the institutional environment, which is the responsibility of the state.

Published in: Facebook, 12 July 2020

More on this topic

Proposals for Developing Leasing in Kazakhstan

Proposals for developing leasing in Kazakhstan, where the KZT 3.3 trillion market is 90–95% state-controlled: accelerated depreciation, removal of VAT barriers, securitization through the AIFC, and support for operating, sale-and-leaseback and real estate leasing. The analysis argues that three changes are most needed: a tax incentive for demand, portfolio refinancing, and the removal of artificial VAT barriers.

On Foreign-Currency Lending

Why Kazakhstan should ease its near-ban on foreign-currency lending: devaluation risk has largely vanished under a free-floating exchange rate, since the interest rate differential (20% in tenge versus 9% in dollars) offsets depreciation. Lifting the 200% capital risk weighting on long-term loans to companies without export revenue and the caps on foreign-currency deposit rates would stabilize funding and let banks lend to the real sector, helping fight inflation.

A 20% Cap on Mortgage Rates Will Cause a Shortage

Why a 20% cap on mortgage rates in Kazakhstan will likely produce a shortage of mortgage lending: banks price loans off the National Bank's base rate rather than their own funding cost, so at this ceiling the product becomes unprofitable for them.