Goldman Sachs: double "signal" for Greek banks from AI and deposits

Summary

Artificial intelligence (AI) is emerging as one of the key structural catalysts for the European banking sector, with Goldman Sachs attempting to quantify the benefits.

According to the US firm, the full adoption of AI applications could boost the return on equity (ROE — Return on Equity, a profitability indicator as a percentage of capital) of European banks by approximately 240 basis points in terms of ROE, without taking into account the initial investment costs.

The analysis is based on 32 real-world use cases and over 100 existing AI initiatives in the European banking sector, covering four key categories: customer service, data analysis, internal process automation, and credit/regulatory control. The main conclusion is that the benefits come mainly from cost reductions and less from revenue increases, with the opportunity on the expenditure side being more than four times that on the revenue side.

What is changing now

In terms of results, Goldman Sachs estimates that the implementation of AI could lead to approximately 2% additional revenue for the industry, with a parallel reduction in operating expenses of approximately 9% and an improvement in risk costs of 7 basis points, through more accurate credit ratings and timely detection of problematic exposures. The combined effect of these factors translates into a potential boost to returns.

However, the benefit is not uniform. Based on different revenue, cost, and capital structures, the difference between banks that are better positioned to leverage AI and those that lag behind can reach 300 basis points in terms of ROE. Banks with higher operating costs, greater labor intensity, and significant exposure to commission and asset management activities show the highest potential benefit.

Greek banks: lower benefit, but also lower risk

For 2027, Greek banks are at the lower end of the distribution, with an estimated ROE boost of between 130 and 170 basis points, compared to an average of 243 basis points for the sample as a whole. This picture reflects the different structure of balance sheets and the fact that Greek banks have already gone through a multi-year phase of consolidation and cost improvement — that is, they have already reaped some of the efficiency gains that AI will bring elsewhere.

At the same time, the report also examines the other side of artificial intelligence: increased competition in commissions and, above all, in deposit costs, due to greater pricing transparency and easier product comparison for consumers.

In a hypothetical scenario of a 25 basis point increase in the cost of retail interest-bearing deposits, Goldman Sachs estimates a negative impact of approximately 3% on pre-tax profits, corresponding to approximately 0.7 percentage points on ROTE (Return on Tangible Equity) on average for the sector.

On this front, Greek banks are among those that show relative resilience, according to the analysis. The majority of deposits in Greek banks are in demand accounts (current accounts with no fixed term) or low-interest accounts, while the share of term products is limited, which reduces the sensitivity of net interest income (NII) to a possible increase in competition on yields.

The balance for the European sector

Goldman Sachs concludes that, on the one hand, artificial intelligence can act as a gradual catalyst for enhancing operational efficiency and returns for the European sector.

On the other hand, it creates new pressures on selected revenues and financing costs. For Greek banks, this balance translates into a modest immediate benefit in terms of ROE from the adoption of AI, but also into a competitive advantage over other European markets in terms of sensitivity to deposit costs.

Watch Now