By Mantvydas Štareika, CEO of SME Bank
Businesses grow by investing. That’s why governments around the world track investment spending so closely. Falling expenditure is an early warning indicator of inadequate growth or even recession.
Funding investment from cash flow is possible but unusual. More often, businesses decide to borrow money to enable the scale of investment they need and to achieve acceptable growth rates.
Getting hold of that money has historically been more challenging for small- and medium-sized enterprises (SMEs). Is that important? Yes, because SMEs are an important part of the European economy, contributing 52.4% of the total value added outside the financial sector.
But access for SMEs to growth finance is stuck in a time warp, where conventional credit scoring risks choking off expansion. This is an unnecessary risk to the health of Europe’s economies as alternative, more accurate methods of credit assessment using AI are now available.
How credit scoring works against startups
Debt financing is hugely important for SMEs. The EU’s latest Survey on the access to finance of enterprises (SAFE), for 2021, finds that across all 27 EU Member States, 77% of SMEs used debt financing in some form. And yet, the same report also finds that accessing business finance becomes more difficult as business size decreases.
One of the many challenges facing smaller businesses looking for investment funds is that lending is often based on the same credit scoring models used for consumer lending. And those models are actually surprisingly basic, covering just basic ID and r previous credit applications and repayment history.
By applying the same models and processes to SMEs, many lending decisions are still being made—and refused—on a relatively thin set of data points that are only available from companies with a few years trading experience. SMEs—and especially Micro SMEs—are often new startups without significant trading histories or the formal bookkeeping in place that legacy banks still demand. Access to investment is harder for startups generally and for SMEs in emerging markets, where lending is still a relatively new concept and a relative lack of banking records.
As a result, lending becomes concentrated on businesses old enough to have an established standard credit history and which operate in more established markets.
AI-based methods of assessing credit worthiness
Many entrepreneurs say they feel left out by banks’ tick-box lending criteria, which fail to take growth potential into account. As a result, many SMEs remain under-financed and at-risk of serious business challenges.
Fortunately there are new, emerging methods of assessing credit risk, based on alternative sources of data currently largely ignored by the credit bureaus. The concept is that AI can make reliable deductions about credit worthiness based on factors that, at first sight, appear irrelevant.
In consumer lending, for example, a number of third party providers are now using AI to look for predictive patterns in the way applicants behave in the apps they use. They are not checking anything that could identify individuals, for example what somebody says in a message or an entry in a calendar. Instead they are checking through the “metadata”—information about information—such as how many entries someone has in their calendar. This metadata, which can be made visible in a safe, permissioned way, reveals important behavioral characteristics. Including propensity to repay loans.
Applying the lessons to SMEs
The same approach is possible in SME Banking too and is gradually being adopted. The approach here can be less to do with things like the number of calendar entries or the completeness of contact data, and more about how consistently data is flowing in and out of accounts. But the concept is the same.
The availability of AI and open banking shortens and simplifies the process of accessing and analyzing data, and introduces unbiased lending criteria derived from contextual trading and cash flow data. There is no reason why a lending decision should take more than an hour. With open banking the data can all be provided remotely and digitally, reducing the bureaucratic burden on businesses when borrowing to expand.
SMEs’ urgent need for financial stability
Ensuring SMEs have access to working capital is recognized as a serious policy challenge by regulators. The European Central Bank reports in its survey on the access to finance of enterprises (SAFE) that, despite a general improvement in finance availability driven by macroeconomic recovery after the pandemic, the smaller the business, the more difficult it is to obtain adequate working capital.
Current economic conditions are just as tough for SMEs as they were during the pandemic. Input prices are increasing for materials and stock, and it is extremely difficult to pass on those cost increases at a time when consumers are squeezed by spiking energy prices and ballooning inflation. This disconnect between costs and income will hit SMEs first and hardest, who typically have less headroom to absorb cash flow shocks, due to inadequate working capital. It is essential they secure their financial stability urgently with new credit lines.
Towards customized scorecards
With the failure of the conventional credit bureaus to devise more inclusive scoring models that reflect the times, progressive banks are now building their own scoring models. The EU has stepped in and supported this approach.
The model under development by SME Finance ecosystem is behavioral rather than based on financial data. AI assesses applicants’ payment behavior, received via the AISP (Account Information Service Provider) provisions of the EU’s PSD2 open banking rules. It can then predict the likelihood a company will fulfill its financial obligations according to the agreed schedule. Applicants who are approved this way get automatic invoice financing. By using artificial intelligence, the model will give smaller companies access to immediate funds and resolve their working capital challenges.
As a policy response, many governments provide credit guarantee schemes for the SME sector, already widespread well before the COVID emergency required emergency measures in this regard. For example, a European Investment Bank (EIB) report in 2017 concluded that “In many Western European countries credit guarantees play a key role in supporting SMEs’ access to finance… In absolute terms, the guarantee sector is largest in Italy (outstanding volume: EUR 33.6bn), France (EUR 16.7bn), Germany (EUR 5.6bn) and Spain (EUR 4.1bn).”
These are national figures and exclude the EIB’s own multinational response, the European Investment Fund (EIF). For SMEs, AI-driven credit scoring is available from some fintechs today. It is likely to become industry standard in the next few years. It is a win-win situation: SMEs, startups in particular, get a better shot at investment loans and banks benefit from fewer defaults.
The ECB’s most recent euro area bank lending survey indicates a considerable tightening of banks’ credit standards has already occurred in the second quarter of 2022. And in the third quarter euro area banks expect an equally severe deflation. The clearer view of credit risk that AI provides will be important in the coming months as the credit squeeze takes effect.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.