Lax monetary conditions leading to low levels of interest rates have been identified as an important driver of risk-taking in the financial sector (see the discussion in Adrian and Liang 2018 as well as the empirical results in Jimenez et al. 2014, and Ioannidou et al. 2015, among many others). In a recent paper (Martinez-Miera and Repullo 2020), we argue that the financial sector's market structure is key in shaping the relationship between interest rates and risk-taking decisions of financial intermediaries.

We consider a one-period risk-neutral economy in which a fixed number of financial intermediaries raise uninsured funding from deep pocket investors and compete à la Cournot in providing loans to penniless entrepreneurs. Intermediaries privately choose the monitoring intensity of their loans, where higher monitoring results in lower probabilities of default. Crucially, we assume that the monitoring decision is costly and unobservable, which creates a standard moral hazard problem between the financial intermediary and the investors. The expected return that investors require for their funds is assumed to be equal to an exogenous safe rate, which can be interpreted as a proxy for the stance of monetary policy.

We show that, in such environment, the effect of changes in the safe rate on the risk of the loan portfolios of financial intermediaries depends on their market power. In competitive loan markets the conventional prediction ensues: lower rates result in higher risk-taking by intermediaries. However, in monopolistic loan markets the opposite holds true: lower rates result in lower risk-taking. We obtain these contrasting results because, although low interest rates lead to low funding costs for intermediaries in both cases, the intensity of the pass-through of financing rates to loan rates depends on their market power, being higher in more competitive markets. Hence, lower safe rates can lead to either lower (in competitive markets) or higher (in monopolistic markets) intermediation margins, which in turn determine lower or higher monitoring incentives for financial intermediaries. We therefore conclude that the underlying market structure is key when assessing the effects of the safe rate on the stability of the financial system.

Figure 1 illustrates this result, showing that an increase in the number of banks leads to a decrease in the slope of the relationship between the safe rate R₀ (on the horizontal axis) and the equilibrium probability of loan default (PD, on the vertical axis). For a sufficiently high number of banks the slope even changes sign, from positive to negative. The conclusion is that market power matters for assessing the effect of interest rates on financial stability. In particular, low interest rates are detrimental to financial stability when banks' market power is low, but beneficial when their market power is high.

**Figure 1** Effect of the safe rate on the the probability of loan default

*Note*: This figure show the relationship between the safe rate and the probability of default for loan markets with 1 (dark blue), 2, 5, 7 and 10 (dark red) banks.

## Alternative competition scenarios

After stating our main results, linking interest rates, market structure, and financial stability, we analyse how our results are affected by three relevant aspects that shape competition in the loan market: (i) the possibility of direct market finance, i.e. financing by investors that (unlike financial intermediaries) do not monitor entrepreneurs, (ii) monitoring cost asymmetries among financial intermediaries, and (iii) entry and exit of intermediaries.

We first consider a situation in which entrepreneurs also have the possibility to obtain direct funding from competitive investors that do not monitor their projects, which in the spirit of Holmström and Tirole (1997) can be related to unsophisticated bond financiers. In such an environment, the equilibrium interest rate that intermediaries can charge is affected by entrepreneurs' outside funding option. In particular, direct market finance imposes a constraint that limits the loan rates intermediaries can charge which reduces their intermediation margins. In particular, we show that monopolistic loan markets exhibit a U-shaped relationship between the safe rate and the equilibrium probability of loan default. In contrast, in competitive loan markets the results of the basic setup do not change. The reason being that in competitive markets direct market finance is not a competitive threat for financial intermediaries (as they already compete intensively among themselves), and therefore it does not affect the Cournot equilibrium outcomes. Figure 2 provides a graphical illustration of these results.

**Figure 2 **Effect of the safe rate on the probability of loan default in the presence of market finance

*Note*: This figure shows the relationship between the safe rate and the probability of default for loan markets with with 1 (dark blue), 2, 5, 7 and 10 (dark red) bank in the presence of direct market finance

We next consider a situation in which financial intermediaries differ in their monitoring costs. We assume that there are two observable types of intermediaries: those with high and those with low cost of monitoring entrepreneurs. In equilibrium, intermediaries with high monitoring costs have lower market shares and their loans have higher probabilities of default. We show that lower safe rates increase (decrease) the market share of high (low) monitoring cost intermediaries and can decrease (increase) the probability of default of their loans. This is the case because lower rates have a higher impact on the margins of high cost intermediaries. We conclude that, in the presence of heterogenous monitoring costs, lower safe rates can have opposite effects on the risk of different intermediaries. We also highlight that, by increasing the market share of those intermediaries with higher cost of monitoring (which grant riskier loans), lower safe rates have an additional ‘composition effect’ on the risk of the financial system, which makes a decrease in the safe rates more prone to increase the average probability of loan default.

We end our analysis of financial market structure by taking into account potential entry and exit decisions of intermediaries. We consider these decisions as a longer run phenomenon compared to the decisions to grant and monitor loans, with the aim of shedding light on the widespread view that interest rates that are "too low for too long" are detrimental to financial stability. We model entry decisions by assuming that intermediaries have to pay a fixed cost to operate. We show that, when entry is taken into account, lower safe rates induce higher competition in the loan market, adding an ‘entry effect’ to our basic results on the effect of low safe rates, which increases risk-taking in the financial sector.

## Alternative funding scenarios

We end our discussion by analysing three alternative funding scenarios for financial intermediaries: (i) replacing uninsured by insured deposits, (ii) introducing competition à la Cournot also in the deposit market, and (iii) funding intermediaries with both (inside) equity capital and uninsured deposits.

Solving the model with insured deposits simplifies the analysis since intermediaries are able to borrow at the safe rate. We show that in this case an increase in the safe rate always leads to an increase in the probability of loan default. The intuition for this result is that, in the perfect competition limit, insured deposits lead to zero intermediation margins and hence zero monitoring, so the relationship between the safe rate and the probability of loan default becomes flat. Away from this limit, i.e. when intermediaries have some market power, lower rates allow them to widen intermediation margins, which translates into higher monitoring and lower probabilities of default. Hence, the results for the model with insured deposits on the effect of safe rates on risk-taking are qualitatively similar to the results for the model with uninsured deposits when banks have significant market power. This highlights the importance of taking into account the composition of intermediaries' funding structure in terms of insured and uninsured debt when analysing the effects of safe rates on the risk of the financial system.

We next consider the effects of changes in safe rates when intermediaries also compete à la Cournot in the deposit market. In this case, we show that the results are qualitatively similar to those of the basic model: low interest rates have a negative impact on financial stability when market power is low, and a positive impact when market power is high.

Finally, we consider what happens when intermediaries can also be funded with inside equity capital, i.e. funds provided by those responsible for the monitoring decisions. As Dell'Ariccia et al. (2014) point out, a relevant determinant of intermediaries' risk-taking decisions is their capital structure, which can be affected by interest rates. We show that when the leverage of financial intermediaries is endogenously determined, market structure can still be a relevant variable in shaping how safe rates affect their risk-taking.

## Conclusion

Summing up, our analysis allows us to conclude that market structure is a key element in the transmission of lower rates to financial intermediaries’ risk-taking decisions. We show that in monopolistic loan markets the pass-through from funding costs to loan rates is weak, hence lower rates result in higher intermediation margins and consequently lower risk-taking by intermediaries. In contrast, in competitive markets the pass-through is strong, such that lower rates result in lower intermediation margins and hence higher risk-taking by intermediaries. This provides a novel testable prediction: Increases in competition (proxied, for example, by the number of banks) reduces the slope of the relationship between the safe rate and the probability of loan default, which changes from positive under monopoly to negative under perfect competition.

Our analysis provides other testable implications regarding the relevance of market structure for the effects of lower safe rates on risk-taking. In particular, when intermediaries' market power is limited by the possibility of firms borrowing directly from investors we predict a U-shaped relationship between the safe rate and the probability of loan default. We also predict that, when banks are heterogeneous in their monitoring technologies, lower safe rates increase the market share of intermediaries with high monitoring costs, a composition effect that moves the overall results in the direction of the competitive benchmark.

Our results also highlight the relevance of certain characteristics of the liability side of the financial intermediaries' balance sheet. In particular, we predict that a higher proportion of insured liabilities (which can be proxied by insured deposits, but might exceed them due to implicit government guarantees) makes it more likely that low safe rates translate into higher intermediation margins and hence lower risk-taking. We also predict that easier access to equity capital (proxied by stock market listing) makes it more likely that low safe rates translate into higher leverage and hence higher risk-taking.

Thus, our theoretical model provides a rich set of novel testable predictions regarding how different market and financial intermediaries' characteristics can affect the relationship between interest rates and risk-taking in the financial sector. However, it should be noted that although we relate the safe rate to the stance of monetary policy, our setup abstracts from other possible relevant effects of monetary policy on aggregate credit demand or deposit supply, which can introduce further relevant interactions left for future research.

## References

Adrian, T and N Liang (2018), “Monetary Policy, Financial Conditions, and Financial Stability”, *International Journal of Central Banking* 14: 73-131.

Dell'Ariccia, G, L Laeven and R Marquez (2014), "Real Interest Rates, Leverage, and Bank Risk-Taking", *Journal of Economic Theory* 149: 65-99.

Holmström, B and J Tirole (1997), "Financial Intermediation, Loanable Funds, and the Real Sector", *Quarterly Journal of Economics* 112: 663-691.

Ioannidou, V, S Ongena and J-L Peydro (2015), "Monetary Policy, Risk-taking, and Pricing: Evidence from a Quasi-natural Experiment", *Review of Finance* 19: 95-144.

Jimenez, G, S Ongena, J-L Peydro and J Saurina (2014), "Hazardous Times for Monetary Policy: What Do Twenty-Three Million Bank Loans Say About the Effects of Monetary Policy on Credit Risk-Taking?", *Econometrica *82; 463--505.

Martinez-Miera, D, and R Repullo (2020), “Interest Rates, Market Power, and Financial Stability”, CEPR Discussion Paper No. 15063.