Nonbank financial institutions are often seen as procyclical lenders, cutting credit more sharply than banks during stress, while bond markets can substitute for bank lending when credit supply contracts. This column brings these perspectives together and argues that this cyclicality depends on the function intermediaries perform. During banking crises, nonbank financial institutions contract syndicated lending relative to banks but expand corporate bond underwriting. For firms with potential access to bond markets, we find that relationships with nonbank underwriters can therefore help preserve market-based financing when bank credit is impaired.
A long-standing idea in corporate finance is that bond markets can act as a ‘spare tire’ when bank lending is disrupted (Adrian et al. 2013, Becker and Ivashina 2014, Cortina et al. 2021). Figure 1 illustrates this pattern. Corporate bonds account for a larger share of firms’ market debt financing during country-specific banking crises, the Global Financial Crisis (GFC), and the COVID-19 shock. At the same time, we document a novel fact: the nonbank financial institution (NBFI) share of corporate bond underwriting is about 7 percentage points higher during banking crises than in non-crisis periods.
Figure 1 Corporate bond shares during periods of financial stress
This raises a question that has received less attention: if firms shift from loans to bonds during banking stress, which intermediaries facilitate that shift? The distinction matters because underwriting differs fundamentally from syndicated lending. Syndicated lenders commit balance-sheet capacity, retain credit risk, monitor borrowers, and may renegotiate loans when firms become distressed. By contrast, bond underwriters certify issuers, build order books, coordinate syndicates, and place securities with investors. Underwriting can involve temporary inventory and reputational risk, but it does not generally require retaining long-run credit exposure. This distinction creates scope for NBFIs active in capital markets, especially broker-dealers, to expand underwriting relative to banks when firms shift from loans to bonds during banking crises. The prediction is therefore not that banks withdraw from bond underwriting, but that NBFIs with stronger distribution capacity gain underwriting share relative to banks. This distinction suggests that a crisis may weaken lending while increasing the relative importance of distribution capacity and investor access.
We study these functions jointly in a new paper (Albuquerque and Firat 2026), using Dealogic primary-market data on syndicated loans and corporate bonds from 1990 to 2025, matched with the updated systemic banking-crisis database of Laeven and Valencia (2026). The sample covers a large set of banking-crisis episodes across advanced and emerging market economies. The data identify both firms and participating lenders or underwriters, allowing us to compare how banks and NBFIs respond within the same borrower and year.
Our first result is that NBFI cyclicality reverses across the two markets. We measure each intermediary’s exposure to a borrower-country banking crisis using the share of its pre-existing loan or bond business located in that country. In specifications comparing intermediaries serving the same borrower in the same year, a one-standard-deviation increase in crisis exposure is associated with 1.3% lower syndicated lending by NBFIs relative to banks (Figure 2). We get the opposite result when analysing the corporate bond market: we find 8% higher bond underwriting by NBFIs relative to banks.
This contrast is consistent with evidence that nonbank syndicated lending is especially cyclical during stress (Aldasoro et al. 2025, Fleckenstein et al. 2026). When funding conditions deteriorate and borrower risk rises, balance-sheet-intensive lending becomes harder to sustain. In our data, the relative NBFI contraction is especially pronounced for term loans. Our novel finding is that the pattern reverses in distribution-based bond underwriting: NBFIs expand underwriting relative to banks when firms seek market debt during banking crises. We also document in the paper that prior nonbank lending relationships are associated with larger loan-spread increases during crises, while prior NBFI underwriting relationships are associated with smaller increases in bond spreads. These results support the claim that nonbank cyclicality depends on the function NBFIs perform.
Figure 2 NBFI response to banking crises across debt markets
After studying how intermediaries adjust their lending activity during crises, we turn to firms’ financing choices during crises. In particular, we ask whether firms with stronger pre-crisis relationships with NBFI bond underwriters are more likely to rely on bonds rather than syndicated loans during banking crises. We measure these relationships using the share of a firm’s bond issuance underwritten by NBFIs over the preceding five years.
In our preferred specification, which compares firms in the same country, sector, and quarter, we confirm earlier evidence that firms are more likely to issue corporate bonds rather than syndicated loans during crises. Our new finding is that this substitution is considerably stronger for firms with pre-existing relationships with NBFI bond underwriters: a one standard deviation higher NBFI underwriting share is associated with a 2.2 percentage point higher probability of a firm switching toward bonds during a crisis (Figure 3). The result survives several alternative explanations. It remains when we control for firms’ prior reliance on NBFI lenders, underwriter reputation, and prior bond-market access, and when we use systemic bank runs from Jamilov et al. (2024) rather than the Laeven-Valencia crisis measure.
The results are also not driven by weak firms being pushed out of syndicated lending and into bonds. Firms with stronger NBFI underwriting relationships are not more leveraged, more vulnerable, or riskier; if anything, they tend to be larger and have lower default risk. Additional tests show that the switching effect remains present among safer borrower groups. Our interpretation is that the ‘spare tire’ may thus not be universal: it is most relevant for firms with potential access to public debt markets and existing underwriting relationships, especially with NBFI underwriters.
Figure 3 NBFI underwriting relationships and switching to bonds
Switching instruments during crises matters only if firms can replace lost loan financing, at least partially. We therefore examine loan issuance, bond issuance, and total market borrowing. Figure 4 shows that during banking crises, syndicated borrowing falls for firms with no prior NBFI underwriting relationship, and it falls even more for firms with stronger NBFI underwriting relationships. But firms with stronger prior NBFI underwriting relationships expand bond issuance substantially more. As a result, total borrowing declines by less (Figure 4).
At the average NBFI underwriting share in our sample, total market borrowing falls by about 6%, compared with roughly 14% for firms with no prior NBFI underwriting relationship. NBFI underwriting relationships therefore offset around half of the decline in new market borrowing. This comparison concerns new syndicated loans and bond issuance rather than all sources of corporate liquidity, such as drawdowns on existing credit lines. With this caveat in mind, the result shows that switching is not only a change in financing composition: it cushions the decline in new external borrowing. NBFIs thus help firms move from balance-sheet-intensive syndicated loans toward market-based debt rather than preserving financing through continued lending.
Figure 4 Borrowing during banking crises and NBFI underwriting relationships
We estimate that NBFI underwriting relationships help preserve financing capacity during banking crises and that this is associated with firms’ relatively stronger financial performance. Firms with stronger pre-crisis NBFI underwriting relationships exhibit stronger tangible and intangible investment growth and smaller increases in their implicit cost of debt during banking crises (Figure 5). By contrast, the effects on employment, sales, liquidity, and default risk are not statistically significant.
Overall, our findings are consistent with the view discussed here that NBFI bond underwriters help mitigate syndicated-loan credit supply shocks by preserving access to market-based debt financing. Although these are reduced-form results rather than estimates of the causal effect of bond issuance itself, they are nonetheless consistent with the idea that access to alternative financing can dampen some of the real effects of banking-sector stress, as in Becker and Ivashina (2014). They also suggest that the identity and function of intermediaries matter not only for where firms borrow, but potentially for how well they weather a disruption in bank credit.
Figure 5 Firm outcomes during banking crises
A final result that we document in the paper helps explain where the increase in NBFI underwriting comes from. We show that it is concentrated among NBFIs affiliated with banking groups rather than independent NBFIs. This points to an organisational margin in the response to banking crises.
Banking groups may use information and client relationships accumulated through lending to support certification and placement by affiliated underwriters. They may also help corporate clients shift toward bond issuance through broker-dealer or securities affiliates when deposit-taking affiliates face tighter balance-sheet constraints. These channels are difficult to separate, but both point to the importance of organisational links between banks and nonbanks. This complements evidence that banking groups reallocate lending toward nonbank affiliates when macroprudential regulation tightens (Albuquerque et al. 2026). During systemic banking crises, however, such reallocation appears more viable in bond underwriting than in balance-sheet-intensive lending. The result therefore does not reflect broad nonbank resilience: the underwriting expansion is concentrated within financial groups.
Our findings suggest that the financial-stability debate around NBFIs should distinguish not only between banks and nonbanks, but also between the functions they perform. The same nonbank sector can amplify stress when it bears credit risk and commits balance-sheet capacity, while helping preserve refinancing capacity when it underwrites and distributes securities. This functional perspective also matters for bank-nonbank interconnectedness: those links can transmit stress, but they can also help redirect financing when one form of intermediation becomes constrained.
Our results do not mean that bond markets provide a complete backstop when bank credit contracts. In fact, public debt access remains concentrated among larger and safer firms, and the underwriting response is closely linked to banking groups. Moreover, assessments of nonbank resilience, regulation, and bank-nonbank interconnectedness should look beyond institutional labels and aggregate credit volumes. In periods of banking stress, the relevant questions are also which function an intermediary performs, where the risk ultimately resides, and whether firms can move across financing markets.
Source : VOXeu
Oil prices fell on Monday as rising Middle East crude exports and a release of…
A trade organization representing community banks sued a US bank regulator Friday, claiming that its…
AI is transforming digital services, and mobile apps, which routinely reach users far beyond their…
Estimates of AI's employment effect disagree in size and even in sign. This column argues…
Military intervention may be undertaken to weaken a hostile regime and strengthen the position of…
Geoeconomic risk has become a first-order concern for investors. This column shows that domestic US…