Small and vulnerable during crises? Firm size and financing constraint dynamics

Heller, David; Karapanagiotis, Pantelis; Nilsen, Øivind A. Article — Published Version Small and vulnerable during crises? Firm size and financing constraint dynamics Small Business Economics Provided in Cooperation with: Springer Nature Suggested Citation: Heller, David; Karapanagiotis, Pantelis; Nilsen, Øivind A. (2025) : Small and vulnerable during crises? Firm size and financing constraint dynamics, Small Business Economics, ISSN 1573-0913, Springer US, New York, NY, Vol. 65, Iss. 1, pp. 451-473, https://doi.org/10.1007/s11187-024-00996-y This Version is available at: https://hdl.handle.net/10419/323679 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1007/s11187-024-00996-y RESEARCH ARTICLE Small and vulnerable during crises? Firm size and financing constraint dynamics David Heller ·Pantelis Karapanagiotis · Øivind A. Nilsen Accepted: 24 December 2024 © The Author(s) 2025 Abstract Thisstudyanalyzesthedynamicsoffinancing constraints under changing economic conditions and the role of firm size in this context. Using administrative data from Germany, we quantify financing constraints expressed as the probability that a firm encounters excess demand or excess supply. On average, smalland medium-sized enterprises (SMEs) are indeed more likely than larger firms to face excess demand for loans. Using the Great Financial Crisis as an empirical setting, we show that tightening financing conditions do not affect smaller firms disproportionally, but generally risky borrowers. Importantly, postcrisis trends in debt-ratios, profitability, investments, Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/ s11187-024-00996-y. D. Heller PolitecnicodiMilano,SchoolofManagement,ViaLambruschini 4b, 20156 Milan, Italy D. Heller (B) Max Planck Institute for Competition and Innovation, Marstallplatz 1, 80539 Munich, Germany e-mail: da[email protected] P. Karapanagiotis University of Groningen, PO Box 72, 9700, AB Groningen, The Netherlands e-mail: [email protected] Ø. A. Nilsen Norwegian School of Economics (NHH), Helleveien 30, 5045 Bergen, Norway e-mail: oi[email protected] and employment are similar irrespective of firm size, while smaller firms respond to the economic slowdown by persistently building up cash buffers. Our results urge policymakers to consider specific characteristics ofbank-dependentfirmstoassess theirexposureto economic crises—instead of focusing on size as vulnerability criteria per se. Plain English Summary During the Great Financial Crisis in 2009, smaller firms tended to build up their cash reserves faster than their larger counterparts. Postcrisis, both SMEs and large firms showed similar trends in debt, profitability, and employment. Thus, the principalimplicationofthisstudy is thatpolicymakersshould focus on specific risk-related characteristics of firms to assess their vulnerability during crises, instead of implementing blanket policies based on firm size. Keywords Financial constraints ·SME financing · Firm-level data ·Bank financing · Small business resilience JEL Classification D22 ·D53 ·G01 1 Introduction Using external financing to leverage business operations facilitates firm-level growth and fosters real economicactivity(Beck etal.,2006;Carpenter &Petersen, 2002; Holmström & Tirole, 1997; Rajan & Zingales, 1998). Financing constraints limit firms’ flexibility to 123 Small Bus Econ (2025) 65:451–473 / Published online: 24 January 2025 D. Heller et al. operate, especially during turbulent market conditions. Smalland medium-sized enterprises (SMEs) rare typicallyconsideredtobeparticularlyexposed to economic downturns (D’Amato, 2020; Eggers, 2020; Vermoesen et al., 2013) due to their inherent informational opaqueness and associated agency problems (e.g., Audretsch 2002; Berger and Udell 2006; Faulkender and Petersen 2006). Consistently, preserving the financing conditions of SMEs during times of economic distress is a common objective pursued by policymakers.1 The Great Financial Crisis of 2009 arguably marks the most severe financial crisis in recent history, urging policymakers and regulators to soften its effects on the most vulnerable firms. The former ECB president Mario Draghi declared that he would do “whatever it takes” to save the European economy and, in particular, its SMEs. More specifically, large-scale programs were initiated to support small firms, such as the ECB’s Outright Monetary Transaction (OMT) program. While these programs certainly fueled the supply of loans (Ferrando et al., 2019; Udell, 2020), they also led to inefficient loan allocation as the policy instruments did not distinguish firms that were in need for help from those that were not (Acharya et al., 2019). Indeed, several studies suggest that financing conditions of SMEs do not disproportionally deteriorate in economiccrisescomparedwithlargerfirms(Atanasova & Wilson, 2004; De la Torre et al., 2008; Kahle & Stulz, 2013;Kremp&Sevestre,2013; Presbitero et al., 2014).2Despite the usual liabilities of smallness, empirical evidence casts doubt on the view to classify SMEs as vulnerable by default: large and small firms adjust their financing patterns during recessions by increasing trade credits (Carbo-Valverde et al., 2016), accumulating cash holdings (Kahle & Stulz, 2013), or reducing investments (Almeida et al., 2009). In this paper, we provide new evidence on the debate about the dynamics of financing conditions of small firms. To this end, we assess how financing constraints evolve over the business cycle, comparing SMEs and large firms. Moreover, we track these firms’ capital 1Figure 6 (Appendix 2) graphically illustrates the increased focus of policymakers on SMEs’ access to credit throughout the Financial Crisis in 2008. 2Aggregates statistics on the Great Financial Crisis in Germany back these findings, indicating that firms’ perceived restrictiveness to obtain bank credits at the outbreak of the crisis was stronger for large than for medium-sized or small firms (Ifo Institute, 2019). structure and other financing variables throughout the Global Financial Crisis. In doing so, this paper sheds light on a set of questions: Do economic slowdowns affect SMEs more strongly than large firms? How do constrained firms adjust their financing and other real economic activities over the cycle? How did the crisis affect the activities and recovery of both SMEs and large firms? Financing constraints are not directly observable such that empirical investigations rely on indirect proxies (Farre-Mensa & Ljungqvist, 2016). Most firmlevel approximations apply to only certain firms, for instance, those that actually pay dividends or have bond ratings (see, e.g., Kaplan and Zingales 1997; Faulkender and Petersen 2006; Whited and Wu 2006). Others, like (Hadlock & Pierce, 2010), are applicable more broadly using firm characteristics such as size and age. These firm-level measures typically have in common that they define being small as a constraint a priori. This attribute creates a challenge for our empirical analysis, regardless of whether it is indeed true that smaller firms are particularly susceptible to financing constraints. To solve this challenge, we use a disequilibrium model based on observable firm characteristics, which enables us to estimate demand and supply for bank loans (similar to, e.g., Atanasova and Wilson 2004; Kremp and Sevestre 2013; Carbo-Valverde et al. 2016). We deploy the estimates of demand and supply to determine the probability of excess demand as a proxy for firms’ degree to be financially constrained. Thus, we assess the degree of being financially constrained in a continuous space such that our results are not dependent on an equivocal threshold that defines whether a firm is constrained (or not). We apply this methodology using administrative data from Germany, the largest economy in Europe. The German economy provides an ideal setup for the research questions at hand because it is traditionally bank-based with close bank-industry relationships (Beck & Levine, 2002; Schmidt & Krahnen, 2004) while having a distinctively strong SME sector (Audretsch & Elston, 2002; De Massis et al., 2018). Combined, the strong dependence on bank financing and its direct use for investment suggests that changes in underlying financing conditions should have measurable effects on German SMEs in terms of borrowing activities. Furthermore, due to its central role in the European economy, implications of adverse shocks to firm-level activities in Germany indirectly affect firms 123 452 Small and vulnerable during crises? Firm size and financing... in other member states, too. Hence, adjustments in firms’borrowingactivitiesandbanklendingpotentially transmitacrossnationalborders.Oursampleconsistsof high-quality, firm-level data from the German Bundesbank, including firm balance sheets, profit and loss, and other descriptive information. The final dataset comprises about 20,000 firms for the years 2006 to 2015, out of which about 75% are SMEs. Our analysis discloses several findings onthedynamics of financing constraints for SMEs and large firms. First, we confirm that SMEs, relative to large firms, are on average indeed more dependent on bank debt, carry higher risk, and have a lower availability of collateral. Second, we find that the probability of being financially constrained is higher for SMEs than for large firms. However, during the economic slowdown, this probability increased between 2008 and 2009 somewhat strongerforlargefirms(7.7%)relativetoSMEs(3.6%). Third, as an immediate response to the recession, all firms, irrespective of size, cut borrowing and investment. Both SMEs and large firms further responded by building up cash buffers, which confirms recent evidence on firms’ mitigation strategies to negative bankloansupplyshocks(see,e.g.,Meinen and Soares2022). Fourth, conditional on being financially constrained, there are very similar recovery patterns for SMEs and large firms, in terms of the use of debt, capital expenditures, and employment. Again, financially constrained SMEs responded to the deteriorating economic conditions by building up persistent cash buffers. This may suggest that SMEs substitute cash for external debt financing,whichisconsistentwiththeoreticalconsiderations, i.e., the pecking-order theory (Jensen & Meckling, 1976), and recent evidence that reports a secular decline in bank lending, in particular of SMEs (e.g., Dell’Ariccia et al. 2021; Falato et al. 2022). The findings of this paper enhance our understanding of the dynamics of financial constraints. We extend prior research on SME financing activities in the German context that uses developments in the 1980s and 1990s as the empirical setting (Audretsch & Elston, 2002; Czarnitzki & Hottenrott, 2011). Our focus is on a more recent economic setting, including the likely most impactful economic crisis since the Great Depression. Further, we contribute to a central stream of the literature on small business economics by investigating small organizations in changing economic environments. Our findings highlight the dynamic organizational capabilities of small businesses discussed in the literature (Nicolas, 2022; Raymond & St-Pierre, 2013; Wai et al., 2022). As we show, SMEs’ operational flexibility helps them to navigate through economic turbulence. Thereby, our analysis contributes to related entrepreneurship literature that examines small firms’ adaptive capacities in changing business environments, such as the digital transformation (Bruque & Moyano, 2007; Escoz Barragan & Becker, 2024;Hu et al., 2023; Kurnia et al., 2015). Unlike these studies, our work demonstrates SMEs’ flexibility using divestments and cash accumulations to soften the adverse effects of recessions. Moreover, our empirical analysis incorporates severaldistinctnewfeatures.Previousanalyses that deploy disequilibrium models in the context of SME financing (e.g., Kremp and Sevestre 2013; Carbo-Valverde et al. 2016) typically define thresholds that classify firms as constrained or unconstrained. Instead, we examine changes in the probability distribution of excess demand or supply across time, i.e., independent of specific cutoffs. Importantly, we include firms of all sizes in our estimations and, therefore, do not restrict ourselves to the perspective of small firms but evaluate the relative impact. This addition allows us to integrate the literatureonthevulnerabilityofSMEs regardingaccess to financing (e.g., Audretsch and Elston 2002;Kremp and Sevestre2013; Vermoesen et al. 2013; D’Amato 2020) into the literature that carves out determinants for small firms’ resilience to changing market conditions (e.g., Almeida et al. 2009; Kahle and Stulz 2013; Nicolas 2022; Escoz Barragan and Becker 2024). We thus extend the literature on SME financing by providing a detailed application of measuring financing constraints and comparing the impact of a significant economic recession across small and large firms. Our findings thus contribute to an important policy debate. European policymakers stress the need to improve access to financing primarily for small firms to stimulate investment and employment. In this context, our results suggest that a more differentiated view on the topic would be appropriate to address the issue of financing constraints effectively. While our results confirm that SMEs are indeed more bank-dependent and have a higher default risk than their larger counterparts, firm size is not equally important in determining financing constraints during recessions. These aspects suggest that policymakers should be most concerned with the underlying causes of what is perceived as the liabilities of smallness, instead of focussing categor123 453 D. Heller et al. ically on firm size, to effectively support small businesses. The paper is organized as follows. Section2presents our data, provides descriptive statistics, and outlines the institutional background. Section3describes our methodological approach in detail. Section4presents the main results, including model extensions. Section5 discusses the implications, generalizability, and limitations of our analysis. Section6concludes. 2 Data and descriptives 2.1 The dataset The empirical analysis is based on proprietary firmlevel microdata provided by the German Bundesbank, the so-called USTAN database. The database contains detailed information on German non-financial firms (e.g., branch, legal form, industry) and their annual financial statements, including profit and loss accounts and the asset history. The USTAN database has been collected annually since 1987 and mainly comprises smalland medium-sized firms. Bundesbank collects these firm-level data within the scope of its refinancing operations, resulting in high-quality, granular data (see Becker et al. (2019) for more details). Table 4 (Appendix 1) lists the original variables (Panel A) and defines the corresponding variables used throughout our analysis (Panel B).3 We use the USTAN database to construct two main datasets. We exclude all firms active in the financial, insurance, and services sectors and those that do not report information on their sector. We also drop observations with below 10,000 Euro of average assets per employee or less than five employees because we cannot reasonably assume them to represent firms that actively participate in financial markets.4For classifying SMEs, its subcategories, and large firms, we follow the recommendation of the European Commission 3Due to significant confidentiality restrictions, it has proven impossible for us to link the firm-level data with bank-level information,suchthat thesupply-sidevariableswillbebased on firms’ balance sheet information. 4Indeed, about two-thirds of the firms with less than five employees in the raw data do not hold any bank loans. Excluding these firms implies that our sample does not comprise many startups but rather well-established SMEs. We reflect on this characteristic in our empirical analysis and the discussion section. (2003/361/EC) on the definition of smalland mediumsized enterprises. For any given year, firms are classified as SMEs once they have less than 250 employees and a balance sheet total of a maximum of 43 million Euros. If they exceed these thresholds, they are classified as large firms. See Table 5 (Appendix 1) for more details on the definitions and the sample distributions. We exclude firms that cannot be uniquely identified as belonging to either SME category. Roughly 75% of sample firms are categorized as micro-, small-, or medium-sized. This number is certainly lower than the actualshare ofSMEs in Germany.However,our sample criteria enable us to preserve the most relevant observations while maintaining a rich enough basis to conduct a thorough analysis. For the main analyses of the impact of the Financial Crisis on German firms, we restrict the time frame to the years 2006–2010, capturing both the boom period before as well as the bust period after 2008. To avoid survivorship bias, we allow firms to enter and leave the database freely as long as they do not enter after 2008.Theresulting dataset covers74,561observations, equivalentto 18,113firms, forthe periodof 5years. The sectoral distributions of firms are displayed in Table 6 (Appendix 1). For the supplementary analysis of firms’ recovery, we construct a second dataset based on the main sample described before. Here, we make two adjustments. First, we extend the sample by adding observations for theyears 2011–2015.More specifically,we adddata for firms observed anytime between 2006 and 2010. As a second adjustment, we only consider firms that are very likely to face financing constraints during the recession (i.e., in 2009) to align the data to assess the recovery of financially constrained firms.5This subsample consists of 4203 firms, corresponding to 50,362 firm-year observations for the period 2006–2015. 2.2 Descriptives Descriptive statistics on several financial items in Table 1provide the first insights on key differences among SMEsand largefirms by comparing basic balance sheet characteristics across firms. In Panel A, several factors 5In particular, we consider both SMEs and large firms that have at least a 75% chance of being financially constrained, according to our disequilibrium estimations in Section 3. 123 454 Small and vulnerable during crises? Firm size and financing... Table 1 Descriptive statistics by firm size category (SMEs vs large firms) Panel A: Comparing basic firm characteristics All SME Large Bankloan ratio 0.222 0.234 0.189 Borrowing diversity 0.372 0.385 0.333 Profitability 2.358 2.536 1.835 Internal funds 0.094 0.103 0.069 Capital expenditures 0.054 0.053 0.057 Acc. receivables 0.296 0.301 0.283 Collateral 0.308 0.281 0.386 Firm size 9.284 8.454 11.725 Public (in %) 7.21 2.80 20.19 Bonds (in %) 3.15 1.03 9.39 Panel B: Composition of the liability side Category All SME Large Bank debt 0.298 0.308 0.268 Bank debt (short-term) 0.160 0.173 0.121 Bank debt (long-term) 0.138 0.135 0.147 Trade credit 0.209 0.211 0.201 Bonds 0.004 0.002 0.012 Nonbank credit 0.183 0.189 0.165 Provisions 0.183 0.163 0.242 Other liabilities 0.123 0.127 0.112 Total 1.000 1.000 1.000 Notes: This table shows mean values of several financial variables in the main sample, displaying values for all firms, SMEs, and large firms in Columns I–III, respectively. All variables are defined in Panels A and B of Table 4 (Appendix 1). In this table, Panel A displays characteristics derived from balance sheet items. Panel B shows the average composition of the liability side of sample firms’ balance sheets. All variables are defined as the balance sheet positions taken from Bundesbank (see Panel A Table 4) as a fraction of total liabilities. Unless indicated otherwise, liability items are aggregated including both shortand long-termed liabilities. Please note that data protection concerns prohibit to display more granular descriptive statistics (e.g., on min max values). Data source: Research Data and Service Centre of the Deutsche Bundesbank (DOI: 10.12757/Bbk.Ustan.8719.05.04), Microdatabase USTAN 2006-2015, own calculations indicate that the average small firm is more dependent on bank debt than its large counterparts. For example, small firms hold higher levels of bank debt, are less likely to be active on capital or bond markets, and have an overall higher concentration on fewer debt sources, as indicated by the borrowing diversity score.6Further, several characteristics indicate that SMEs are likely to be riskier borrowers than large firms. The level of risk can be inferred from slightly higher levels of accounts 6We compute this index, following (Tengulov, 2019)which measures the concentration of different debt types. Larger values reflect a higher dispersion in debt sources and, thus, a higher concentration on fewer financing sources. Table 4 (Appendix 1) lists and defines all variables. receivable (as an indicator for the ability to collect payments), lower availability of collateral (as measured by tangible assets), which is reflected in higher operating risk, and lower holdings of tangible assets (i.e., potentialcollateral). All differences mentioned above are statistically significant at the 1% level, as (unreported) t-tests show. Regarding financing needs, descriptives show similar capital expenditure levels among SMEs andlargefirms. Thesestatistics indicatethatsmallfirms are typically on a different stage of the firm life cycle, which is confirmed by the differences in the mean age and higher growth rates, i.e., profitability rates. All differences are statistically significant at the 1% level (unreported). 123 455 D. Heller et al. In Panel B, we take a closer look at the composition of the liability side of firms’ balance sheets, using liability ratios of different balance sheet items. We observe smaller firms to have a much higher short-term borrowing (17.3 versus 12.1% of liabilities), whereas larger firms obtain more long-term debt. Overall, however, the share of total bank debt is higher for small firms.Moreover,a muchlowerfractionof bondsresembles the quasi-absence of small firms in those markets. These stylized descriptives mirror the limited number of financing sources and stronger dependence on bank financing of small firms, as proposed in previous literature (e.g., Faulkender and Petersen 2006). Especially when considered in combination with the characteristics displayed in Panel A, statistics from Panel B illustrate why smaller firms are commonly assumed to be more dependent on bank financing and, thus, more vulnerable to exogenous shocks in the loan supply. 2.3 Institutional background: the financial crisis in Germany In the following, we outline the institutional context of our empirical analyses. Shortly after the financial crisis shocked financial markets in the US at the end of September 2008, it spilled over to the European financial sector. Because of the global involvement of German financial institutions in the global credit market, the German financial and credit system began to falter, eventually forcing the state to launch a 500 billion Eurosrescue programfor thefinancial industryin October 2008. Subsequently, demand for German industrial products and international financial markets collapsed massively, slowing GDP growth to 1.0% in 2008 and a slumping −5.7% in 2009. However, business expectations recovered from an all-time low, indicating a positive outlook by the first quarter of 2010 (Ifo Institute 2019). An important aspect of our analysis is the effect of the Financial Crisis on bank lending in Germany. Figure1illustrates the perceived bank lending conditions at the time by plotting the firm size-specific Ifo Credit Constraint Indicator (Ifo Institute, 2019). The graph documents the fraction of firms perceiving the current willingness of banks to extend credit as restrictive. During the periods of 2005 until 2008 and 2011 until 2012, the perceived degree of being financially constrainedamongallfirmswasat a relativelylowlevel of 20% overall but more than doubled in 2009.7In line with common perceptions on financing constraints of small firms (e.g., Berger and Udell 2006), before 2009, smalland medium-sized firms were reportedly more constrained than larger firms. More importantly, however, the increase in perceived credit constraints at the outbreakoftheFinancialCrisisin2009is mostsubstantial for large firms, despite the higher initial values of the other two size categories. These observations contrast the common perception that smaller firms are hit particularly strongly by adverse economic effects (e.g., Holmström and Tirole 1997; Eggers 2020). Further, these observations raise the question of whether small German firms were indeed disproportionally affected by the Financial Crisis and—if not—how did they fend off the adverse economic shock? 3 Methodology 3.1 Considerations on financing constraints Financing constraints are typically not observable, but several measurement approaches exist. Most measures consider firm size qua definition or are only available for public firms. This is problematic for our analysis because we examine firms of different size categories, includingmostlyprivate small firms, as outlined in Section2.2.Anotherkeychallengeforestimatingfinancing constraints is disentangling supply and demand effects. To solve these two aspects, we deploy a disequilibrium method as implemented, e.g., in Kremp and Sevestre (2013), Carbo-Valverde et al. (2016), and described in Karapanagiotis (2024). This approach is advantageous for several reasons. First, integral to our research question, it does not classify small firms as constrained per se, and it can be estimated for any firm. Second, unlike studies that explore information on loan applications via credit registries (e.g., Cantú et al. 2022), this approach allows analyzing demand effects, including information on discouraged borrowers (i.e., those who did not apply for loans). Third, our approach does not rely on survey data to delineate demand and supply factors (as in Presbitero et al. (2014)). However, surveys typically rely on specific selections and may not include sufficient financial data. 7To illustrate that this observation is not source-specific, Fig.7 (Appendix 2) displays an alternative descriptive statistic from the Deutsche Bundesbank’s bank lending survey. 123 456 Small and vulnerable during crises? Firm size and financing... Fig. 1 The Financial Crisis in Germany: Financing climate between 2006 and 2012.Notes:The figure graphically illustrates the responses to the Ifo Credit Constraint Indicator for manufacturing firms. To calculate the intensity of credit constraints, firms are asked to answer the question: “How would you assess the current willingness of banks to extend credit to businesses?” Respondents choose between the answers “accommodating,” “normal,” and “restrictive.” Own illustration based on data from (Ifo Institute, 2019) The disequilibrium model avoids these caveats by estimating credit demand and supply based on firmlevel financial data. Moreover, it provides us with a continuous measure of the intensity of financing constraints expressed as the probability of a firm experiencing excess demand or supply.8Hence, our approach conceptually distinguishes demanded and supplied quantities (Qdand Qs), described in the following linear system: Qd it =Xd it βd+X itαd+εd it, Qs it =Xs itβs+X itαs+εs it, (1) of firm iat time t, where Xk it (for all k∈{d,s}) denotes vectors of explanatory variables in the demand and the supply equations. The Xit is a vector of common controls in the two equations. The shocks are assumed to be 8Temporary deviations that lead to excess demand or supply in the loan market can be caused by costly information gathering and processing and by the menu costs of banks (Mackowiak & Wiederholt, 2009; Mankiw & Reis, 2002) As such, information asymmetries impede a proper assessment of whether changes in the demand for loans are temporary, idiosyncratic, or uniform to the whole market. Moreover, menu costs induce banks to define a target retail rate as a function of long-term market interest rates. bivariate normally distributed with εd it ∼N(0,(σd)2), εs it ∼N(0,(σs)2), and where the correlation of εd it and εs it is ρ(which we might set equal to zero). Our approach combines the equations of the system Eq. 1 with the following condition: Qit =min{Qd it,Qs it},(2) where Qit is the observed amount of loans that resembles the minimum of demanded or supplied loans. This expression implies that the observed quantity belongs either to the demand side or the supply side in Eq.1. Since this fact cannot be directly observed, the model uses a log-likelihood method to predict hypothetical quantities for both demand and supply ( ˆ Qd it and ˆ Qs it). In other words, the log-likelihood method where Eqs.1 and 2are estimated simultaneously gives for each observation in our panel data, a probability whether it belongs to Qdor Qs. Based on the output of these estimations, we construct two measures for the intensity of financing constraints. For the first measure, we start with the follow123 457 D. Heller et al. ing expression: θit = ˆ Qd it −ˆ Qs it ˆσ2 d+ˆσ2 s−2ˆσdˆσsˆρ .(3) The θit is the expected excess demand of firm iin period t. In other words, θit is a non-binary measure of the intensity of financing constraints, which we normalize by the standard deviation of the difference of the demandandsupplyshocksestimatefor all observations in the sample. With this expression at hand, we then calculated the probability of resembling a financially constrained firm in a given year for each observation following Maddala and Nelson (1974, Eq. 2.2). The probability of an observation representing a financially constrained firm is equal to the probability of the firm being in an excess demand regime (with denoting the standard normal distribution), i.e., ˆπd it =P(Qd it >Qs it)=(θit).(4) Our second measure specifies a statistic value of excess demand by the difference between expected excess demand and supply relative to the expected supplyforloans.Thismeasurecapturestheextentof excess demand, i.e., Fit = ˆ Qd it −ˆ Qs it ˆ Qs it .(5) In Fit, the estimated excess demand is normalized by expected supply, which is observation-dependent and, thus, idiosyncratic. Just like θit,Fit is independent of the unit of measurement. By regressing Fit onaset of independent variables, we can investigate whether size characteristics are indeed relevant in explaining a potential increase in financing constraints throughout the financial crisis. 3.2 Variable specifications and estimation results of the disequilibrium model As the main dependent variable, Qit, we specify firms’ bank-loan ratios as the end-of-the-period total amount of loans owed to credit institutions as a fraction of total assets. In the following, we define determinants for the demandand supply of loans. To allowforidentification, we have to make a qualified selection of variables in each of the two equations, i.e., imposing an exclusion restriction (see, e.g., Carbo-Valverde et al. 2016). Table 4 (Appendix 1) contains an overview of all variables of our empirical model. Demand for bank financing is particularly affected by its price and the availability of other, less expensive substitutes because external debt financing is a relatively costly funding option (Love et al., 2007; Myers & Majluf, 1984). We, therefore, model loan demand as a function of the interest rate and financing alternatives.9Forthefinancing alternatives,weconsider firms’ internal funding via cash flows. Further, we consider firms’ capital expenditures as a more direct measure of firms’ (long-term) financing needs. Moreover, firms with diversified debt structures are less vulnerable to economic slowdowns (Giannetti, 2019), which is why we incorporate a measure of firms borrowing diversity. The supply-side determinants relate to default costs expressed as the risk associated with the potential borrower (Freixas & Rochet, 2008). We thus follow related studies, such as (Kremp & Sevestre, 2013), and consider three variables that reflect less risky and less volatile cash flows, all of which should translate to a higher willingness of banks to supply loans: (i) the share of accounts receivable to total assets as an indicator of firms’ ability to collect payments on their business activities (i.e., lower liquidity risk), (ii) the availability of collateral as measured by firms’ stock of tangible assets, and (iii) profitability, as measured by the returns to assets.10 Further, we acknowledge that several covariates affect loan demand and supply simultaneously. First, firm size helps to control for the firm-specific risk and the need for bank debt, thereby addressing both loan demand and supply factors. Second, controlling for time fixed effects (year-fe) is important because it allows, among others, to control for the refinanc9We impute a hypothetical interest rate for firms with zero loans using a coarsened exact matching (CEM) procedure for observations without outstanding bank loans. This applies to 24% of our sample firms. Appendix 3 elaborates on this in detail. For robustness, our analyses contain results for firms with and without imputed interest rates. 10 Further, our specification incorporates the fact that banks set prices according to borrowers’ risk and not demanded quantities, i.e., banks first decide how much they are willing to lend before they set interest rates (Kremp & Sevestre, 2013). We follow this consideration and do not include interest rates as a supply-side determinant for lending. 123 458 Small and vulnerable during crises? Firm size and financing... Fig. 3 Short-term adjustments to tightened financing constraints, SMEs vs large firms.Notes: The figure plots coefficients of three separate regressions, each of which estimates Eq.7. Estimations use three different samples: (1) full sample, (2) SMEs, and (3) large firms. The graph plots the respective βcoefficients for each of the subsamples. The dependent variables are bank-loan ratios (A), the borrowing diversity index (B), and the cash and cash equivalents to the total asset ratio (C). All variables are defined in Table 4 (Appendix 1). 2006 is the reference year. Whiskers represent 95% confidence intervals. Standard errors are heteroscedasticity-consistent and clustered at the firm level. Data source: Research Data and Service Centre of the Deutsche Bundesbank (DOI: 10.12757/Bbk.Ustan.8719.05.04), Microdatabase USTAN 2006–2015, own calculations tive and significant, whereas it is insignificant for large firms. Notably, however, the overall development of an increased concentration on fewer debt sources already unfolded in 2008, whereas the crisis mainly affected German firms in 2009. Hence, the concentration on fewer financing sources may also reflect the flexibility ofthe firms to react early on to the declineinbusiness.13 Furthermore, we investigate whether SMEs actually mitigate the adverse shock of the Global Financial Crisis by prudent behavior, i.e., by adjusting their holdings of cash and cash equivalents. Firms, and in particular smaller ones, are found to be more likely to be liquidated when they are in financial distress (Ozkan & Ozkan, 2004), which might induce them to increase cash holdings to avoid financing distress in the first place. This precautionary effect may occur in response to a sudden shock (e.g., Dessaint and Matray 2017). Panel C shows that firms indeed strongly increased cash holdings in 2009. The effect is highly significant and translates roughly to a 20% increase in cash for the median SME. The effect holds for both SMEs and large firms; higher associated costs of financial distress can explain why firms build up cash buffers as a reaction to the overall slowdown in economic activity (Kahle & Stulz, 2013). Moreover, we find that both small and large firms decrease their capital expenditures with the onset of the Financial Crisis in a very similar manner (see Fig.9B in Appendix 2). Notably, also the impact on capital expenditures appears virtually equivalent for small and large firms. Against the background of our previous results, this suggests that firms substitute external funding by building up cash buffers. 4.4 Longer-termed implications Whereas our previous analyses focus on the initial impact of the Global Financial Crisis on firms’ financial positions, it is reasonable to assume that some effects—including recovery processes—are only visible from a longer-termed perspective. In the following, we extend our analysis by studying the post-crisis 13 For example, the data mirror aggregate statistics from the Ifo Business Climate Index, which separately reports business expectations and the current business situation (see Fig.9A in Appendix 2). Although the current business situation only declined in 2009, expectations already worsened as of the second half of 2008. 123 465 D. Heller et al. period until 2015. More specifically, we focus on those firms identified by the disequilibrium estimations as being financially constrained during the recession year 2009. If financing constraints during crises lead only temporarily to worse performance, these firms should have recovered with the economy recovering as well. Hence, analyzing firms that faced financing constraints in 2009 and distinguishing between SMEs and large firms will provide a more complete picture of the implications of the crisis on these firms. Therefore, we extend our main sample by adding observations for 2011–2015 as described in Section 2. Again, we consider SMEs and large firms whose probability of being financially constrained as defined in Eq. 5is equal to or larger than 75%. This way, 4203 individual firms, or 28.1% of all firms in the sample, are flagged as financially constrained in 2009. The fraction of SMEs that classify as constrained (31.0%) is higher than the fraction of large firms (18.7%), which is consistent with our previous analysis, e.g., Fig.2.Using the extended sample, we estimate the long-term evolution in financing and other economic activities for financially constrained SMEs and large firms by reestimating Eq.7. Moreover, we use two sets of outcomes as dependent variables that relate to the (a) financing and (b) real economic activities of firms, which are summarizedin Figs.4and5,respectively. Similar to before, we plot the yearly coefficients from the event-study-type regressions. Regarding financial activities, in Fig.4, we assess three specific outcomes. First, in Panel A, we usefirms’ bank-loan ratios as the dependent variable and observe the deleveraging trend over the entire extended sample period. This observation is consistent with recent empiricalevidencefrom SMEsin the USthat findlower use of debt financing as a persistent long-term trend since the 1980s (e.g., Dell’Ariccia et al. 2021). However, we find no statistically significant difference in this trend between SMEs and large firms, suggesting that the deleveraging applies equally to firm size categories. Second, in Panel B, we investigate borrowing diversity: During the crisis years, firms appear to concentrate increasingly on fewer sources of external financing, including the previously observed anticipatory effect in Section 4.3; at the latest by 2013, firms revert to pre-crisis levels of borrowing diversity. As an important finding, this is again very similar among constrained SMEs and constrained large firms. Consistent with Giannetti (2019), this finding suggests the crucial role of debt concentration during times of economic distress. Third, in Panel C, we use the cash-to-liability ratio as the dependent variable. SMEs increased their cash holdings with the onset of the crisis in 2009. In the aftermath of the crisis, they initially lowered these amounts slightly but increased cash holdings thereafter. This pattern is different from larger firms: They increased cash holdings in 2009, but unlike smaller firms, large firms reduced these cash holdings by 2011 to pre-crisis levels and maintained those until 2015. ThesefindingssuggestthatfinanciallyconstrainedGerman firms respond to crises by building up cash buffers to strengthen their financing position. This observation would be in line with other studies on the response of firms and managers to adjust financing positions in response to adverse shocks (e.g., Kahle and Stulz2013; Dessaint and Matray 2017). On top of this, SMEs build up these buffers even more persistently. Combining the three findings of sustained deleveraging, an increase in cash holdings, and no change in borrowing diversity in the longer term suggests that firms, and in particular SMEs, potentially substitute cash for debt. As a next step, we investigate the evolution of financially constrained firms’ real economic activities, measured in terms of firm profitability, capital expenditures,andemployment.Figure5showsa consistentpicture across these three dimensions; that is, the average constrained firm exhibits lower levels of real activities during the crisis period.14 Moreover, both constrained SMEs and large firms in our sample recovered with increased profitability immediately after the crisis in 2011 (Panel A). Similarly, employment rates reverted to precrisis years immediately after the crisis (Panel B). However, some of the longer-term implications differ across firms. As such, Panel C shows differences in the recovery regarding capital expenditures. Constrained SMEs reverted to precrisis levels by 2010. Although this reversal is persistent over time, the level of investment of constrained large firms remains at the lows of the crisis. For robustness, we back all of these obser14 Regarding potential firm failures, we observe that 24.2% of SMEs and 23.7% of large firms active in 2007 dropped out of the sample by 2012. However, the mere fact that firms are still alive does not rule out weak performance. As such, it has been shown that many firms survive crises but remain in a zombie-like state in the aftermath of economic slowdowns (e.g., Acharya et al. 2019). 123 466 Small and vulnerable during crises? Firm size and financing... Fig. 4 Evolution of financing activities by constrained SMEs and large firms (2006–2015).Notes: The figure displays the evolution of changes in financing activities differentiating between financially constrained SMEs and constrained large firms. The graphs plot coefficients (β) similar to those on estimations of Eq.7, only here we use an extended sample, covering the years 2006–2015. The dependent variables are specified as before: Bank-loan ratio (A), borrowing diversity (B), and firms’ cash holdings, measured as cash-to-liability ratio (C). The coefficients plotted are for each year, relative to 2007. Whiskers represent 95% confidence intervals. Standard errors are heteroscedasticityconsistent and clustered at the firm level. Data source: Research Data and Service Centre of the Deutsche Bundesbank (DOI: 10.12757/Bbk.Ustan.8719.05.04), Microdatabase USTAN 2006-2015, own calculations 123 467 D. Heller et al. Fig. 5 Long-term impact on real activities of constrained SMEs and large firms (2006–2015).Notes: The figure displays the evolution of changes in real economic activities similar to Fig.4. The dependent variables are profitability measured as EBIT over totalassets(A),investmentmeasuredascapitalexpendituresover total assets (B), employment measured using the logarithm of the number of total employees (C), and firm-level markups measured by dividing total operating income by operating expenses (D). The coefficients plotted are for each year, relative to 2007. Whiskers represent 95% confidence intervals. Standard errors are heteroscedasticity-consistent and clustered at the firm level. Data source: Research Data and Service Centre of the Deutsche Bundesbank, Microdatabase USTAN 2006–2015, own calculations vations by testing alternative specifications of the measures for profitability, growth, and capital expenditures (see Fig.10 in Appendix 2). Summarized, constrained SMEs and large firms appear to respond similarly at the onset of the crisis, a pattern confirming aggregate statistics introduced in Section 2. On top of this, we find recovery patterns to be comparable. If anything, the recovery of SMEs is stronger, in particular regarding investment rates in the form of capital expenditures. As a final step, we also investigate changes in the markups of respective firms. According to Meinen and Soares (2022), firms that are more exposed to liquidity risks, especially financially constrained firms, tend to raise markups in response to adverse bank-loan supply shocks to sustain liquidity. We measure markups as the share of income from operating activities over operating expenditures. Figure 5D again plots estimates of β coefficients in estimating Eq.7, this time using firmlevel markups as the dependent variable. Using firmlevel markups as a dependent variable shows that SMEs indeed increased markups during 2009 and 2010 while reverting to 2006 levels from 2011 onwards. This pattern is well in line with Meinen and Soares (2022). However, we do not find statistically significant patterns like this for larger firms. The above evidence strongly suggests that SMEs sustained the negative impact of the Global Financial Crisis by increasing markups, short-term divestment, and a sustained accumulation of cash holdings. Whereas these patterns apply also to larger firms, they 123 468 Small and vulnerable during crises? Firm size and financing... are most pronounced for SMEs. These findings highlight potential mechanisms for how small firms can cope with adverse economic shocks. Moreover, among several measures, we do not detect significant differences in other financing activities or performanceand growth-related outcomes between SMEs and large firms. 5 Discussion 5.1 Contributions and implications Our analyses provide new insights into a central question in the literature of small business economics: How do small firms cope with changes in their business environment, and to what extent are they capable to sustain adverse economic shocks? The traditional view on the liabilities of smallness posits that smaller firms are resource-constrained, especially in terms of financial assets and skilled personnel (Aldrich & Auster, 1986; Freeman et al., 1983). A common perception is that these liabilities leave small firms more vulnerable to slowdowns in economic activity (see Holmström and Tirole 1997; Faulkender and Petersen 2006; Eggers 2020). Our work corroborates the fact that SMEs were significantly hit by the financial crisis (see, e.g., Udell 2020). Importantly, these effects were not disproportionally more severe for SMEs than for large firms in Germany. Our results also suggest that SMEs dynamically adjust their operations in response to the crisis. In particular, these adjustments include short-term divestment that yields cash accumulation as well as increasing markups. Our study contributes to empirical and theoretical considerations on the role of small organizations in changing economic environments. In particular, a prominent stream of related literature describes the dynamicorganizationalcapabilities ofsmall businesses (Nicolas, 2022; Raymond & St-Pierre, 2013; Wai et al., 2022).Here, the term smallness is not only linkedto liabilities but also to capabilities, such as operational flexibility. As such, the organizational structures of small firms are less complex and hierarchical than those of larger peers, implying better abilities to adapt to changing business environments. In this context, existing work often focuses on different forms of technology adoption (Bruque & Moyano, 2007; Escoz Barragan & Becker, 2024; Hu et al., 2023; Kurnia et al., 2015). In contrast, our analysis contributes to the literature by demonstrating the flexibility of SMEs to adjust operations and their underlying workings during times of economicturbulence.Thereby,weofferanuancedview on the traditional perspective of the liabilities of smallness as we show different mechanisms for how small firms can navigate through times of economic distress. These insights have several important practical and policy implications. They contribute to the discussion of how new small firms can overcome resource constraints, especially during turbulent times. Our findings help to gain a better understanding of the underlying workings of how small firms can overcome their liabilities of smallness, offering guidance to decision-makers inSMEs: We suggest managers tofocuson the dynamic capabilities of their firms in order to sustain economic turbulences more effectively. Most directly, we contribute to an ongoing debate among policymakers and researchers on the vulnerability of SMEs (Audretsch, 2002; Eggers, 2020). Policymakers and academics have increasingly focused on the financing conditions of small firms. For example, Fig.6 (Appendix) shows that about every other speech by respective ECB presidents in the aftermath of the Global Financial Crisis discussed SMEs’ access to credit. Our insights into this debate are crucial, given the significant policy initiatives launched in the advent of the crisis. In his infamous speech, the former ECB president Mario Draghi announced he would do “whatever it takes” to safeguard the European economy and, in particular, its SMEs. As a consequence, in the US and in Europe, regulators initiated massive programs in support of small firms, such as the Outright Monetary Transaction (OMT) program by the ECB or the Troubled Asset Relief Program (TARP) in the US. Research has shown that these programs stimulated credit supply, softening SMEs’ financing constraints both in the US and Europe (e.g., Li 2013; Ferrando et al. 2019; Udell 2020). Unsurprisingly, these large-scale programs stimulated access to bank financing and stabilized the banking sector as a whole. Yet, most studies remain silent on the efficiency of the respective programs. In fact, evidence suggests that the ECB’s OMT program led to credit misallocations that slowed economic recovery (see Acharya et al. 2019). Our find123 469 D. Heller et al. ings corroborate this perspective, suggesting that policymakers should place more emphasis on specific characteristics of bank-dependent firms to define potential exposures to crises instead of mere size delineations. This aspect is particularly important given the strong focus of policymakers on designing structural help programs using merely size-based measures. We thus provide new insights into the underlying reasons why the massive policy initiatives in the aftermath of the Global Financial Crisis lacked precision. 5.2 Generalizability, limitations, and future research Financial crises mark extraordinary periods that can fundamentally affect firm dynamics. Our empirical analysis relies on one specific crisis and one specific set of firms—the Great Financial Crisis and German SMEs. Still, its implications are likely generalizable to other crises and other types of firms. The results of our analysis are likely applicable to other crises, including equally significant shocks like the COVID-19 pandemic. In general, recessions typically hit firms with a sudden decline in demand for their output. Under these conditions, firm revenues typically drop, while their obligations to cover operating costs and to serve creditors and suppliers remain (see, e.g., Gourinchas et al. 2020). In what follows is that economic activity is shaped by uncertainty until demand eventually reaches a pre-crisis level. Along these lines, the Great Financial Crisis is not notably different to other crises, such that the general workings of how SMEs react to it are likely similar. What is different, however, is the triggering circumstances and the severity of the crisis. Important in terms of generalizability is the magnitude of the Global Financial Crisis. In this regard, our settings stand out as we focus on a particularly harsh crisis (see Section 2.3). Our results highlight the resilience of German SMEs during these times, suggesting that they should sustain more moderate recessions as well. In a similar vein, the findings of our paper likely apply to SMEs outside of Germany, too. Studying German SMEs is advantageous given the importance of banks as a primary source of external financing for investments. Similar to many other countries, in Germany, the SME sector is of central importance for the country’s economic activity (Audretsch, 2002; Audretsch & Elston, 2002; De Massis et al., 2018; Eggers, 2020). Likewise, the characterizing features of SMEs in Germany (i.e., opacity and bank dependence) do not differ from those of SMEs in other countries. Overall, our results are, therefore, indicative for other settings—both in the past and in the future. Given the frequency in which firms undergo economic turbulences, this paper contributes to a central question in the literature of small business economics. Nevertheless, our analysis remains subject to certain limitations. First, like all data-driven approaches, the generalizability of our empirical results has natural limits. Our study captures SMEs from one country, implying that industry characteristics or the regulatory environment may cause the implications to differ when considering other economies, in particular, those outside of Europe (see Udell 2020). As such, the European Sovereign Debt crisis that unfolded in 2010 for a subset of Euro-area countries, excluding Germany, followed the financial crisis. It triggered a series of effects within respective countries that likely differ from those observed in Germany. Moreover, our analysis captures the financing dynamics of already established SMEs to the crisis. In addition to this, it would be interesting for future research to focus more specifically on startups, such as entrepreneurial ventures, and on other operational adjustments, like innovation-related investments. As another specificity, our empirical analysis leverages highly detailed and reliable financial information from administrative sources. The level of granularity enables us to dissect different dimensions of firm-level financial activities to an extent that is typically not possible with ready-to-use balance sheet information. Still, the high quality of the data comes at the cost of having relatively limited information on the corresponding banks, i.e., the financing supply side. Similarly, the data underrepresents very small and young firms as they are oftennotsubjecttotherigorousreportingrequirements. An economy-wide mapping of bank-firm links at the level of granularity of our data would certainly help future research to further advance our understanding of SME vulnerability in times of economic crises. 6 Conclusion SMEs contribute significantly to economic growth and prosperity. However, these firms are often considered particularly prone to adverse economic shocks—an 123 470 Small and vulnerable during crises? Firm size and financing... observation questioned by both anecdotal and previous empirical analysis. This paper investigates the dynamics of financing constraints, focusing on the role of firm size in the German economy. Germany is historically considered a bank-based economy and, thus, suggests that the scarcity of funding opportunities stemming fromthe banking sector is a good indication of an actual worsening of firms’ ability to obtain financing. Theanalysis showsthatSMEs are not disproportionally affected by the banking crisis. Indeed, before and during the crisis, the average SME exhibited a higher probability of being financially constrained. However, we find robust evidence that the effect of the crisis on financial constraints is not significantly greater for small firms than for large ones. In the post-crisis years, we observe a deleveraging trend for both financially constrained SMEs and constrained large firms. Regarding the longer-term implications of the slowdown in economic and financing activities, we find that SMEs persistently build up cash buffers, potentially substituting cash holdings for debt. Further, both financially constrained SMEs and financially constrained large firms recovered from the crisis similarly by regaining profitability, investments, and growth. Our results highlight that firms’ size cannot determine the intensity of financial constraints per se. Instead, non-bank funding alternatives and internal funds are most important in mitigating financial constraints. The analyses provide new evidence on the vulnerability of small, bank-dependent firms in the context of the largest banking-based economy worldwide. Acknowledgements We especially thank two anonymous referees and editor Julie Ann Elston for numerous comments and suggestions. We are also thankful for fruitful discussions and comments by seminar participants at Copenhagen University, Goethe University Frankfurt, Norwegian School of Economics, Annual Congress of the Verein für Socialpolitik (German EconomicAssociation) 2020, the FourthERMEES Macroeconomics Workshop 2021, the 11th RCEA Money, Macro, and Finance Conference 2021, International Risk Management Conference 2022, and Royal Economic Society Annual Conference 2023 for theircomments andsuggestions. We thank theResearch Dataand Service Centre (RDSC) of Deutsche Bundesbank in Frankfurt for their hospitality and support. The project number at Bundesbank is 2018\0051 and the data was obtained via a secure on-site access (Gafo). The article reflects the opinions of the authors only and does not express the view of Deutsche Bundesbank nor should it be attributed to the institution. All remaining errors are our own. Funding Open access funding provided by Politecnico di Milano within the CRUI-CARE Agreement. David Heller received funding by the European Union – NextGenerationEU, Mission 4, Component 2, in the framework of the GRINS – Growing Resilient, INclusive and Sustainable project (GRINS PE00000018–CUPXXXXX). Theviewsandopinionsexpressed are solely those of the authors and do not necessarily reflect those of the European Union, nor can the European Union be held responsible for them. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Acharya, V. V., Eisert, T., Eufinger, C., & Hirsch, C. (2019). Whatever it takes: The real effects of unconventional monetary policy. The Review of Financial Studies, 32(9), 3366– 3411. Aldrich, H. Auster, E. (1986). Even dwarfs started small: Liabilitiesofageandsizeandtheirstrategicimplications.Research in Organizational Behavior/JAI Press. Almeida, H., Campello, M., Laranjeira, B., Weisbenner, S. (2009). Corporate debt maturity and the real effects of the 2007 credit crisis. National Bureau of Economic Research, 14990. Atanasova, C. V., & Wilson, N. (2004). Disequilibrium in the UK corporate loan market. Journal of Banking & Finance, 28(3), 595–614. Audretsch, D. B. (2002). The dynamic role of small firms: Evidence from the US. Small Business Economics, 18, 13–40. – and Elston, J. A. (2002). Does firm size matter? Evidence on the impact of liquidity constraints on firm investment behavior in Germany. International Journal of Industrial Organization, 20(1), 1–17. Beck, T., Demirgüç-Kunt, A., Laeven, L., & Maksimovic, V. (2006). The determinants of financing obstacles. Journal of International Money and Finance, 25(6), 932–952. – and Levine, R. (2002). Industry growth and capital allocation: does having a market-or bank-based system matter? Journal of Financial Economics, 64(2), 147–180. Becker, B., & Ivashina, V. (2014). Cyclicality of credit supply: Firm level evidence. Journal of Monetary Economics, 62, 76–93. 123 471 D. Heller et al. Becker, T., Biewen, E., Neulen, R., Schultz, S., Weisbecker, M. (2019). Corporate Balance Sheets (USTAN) 1999-2015. Data Report 2019-01 Metadata Version 1., Deutsche Bundesbank Research Data and Service Centre. Berger, A. N., & Udell, G.F.(2006). A more complete conceptual framework for SME finance. Journal of Banking & Finance, 30(11), 2945–2966. Blackwell, M., Iacus, S., King, G., & Porro, G. (2009). Cem: Coarsened exact matching in stata. The Stata Journal, 9(4), 524–546. Bruque, S., & Moyano, J. (2007). Organisational determinants of information technology adoption and implementation in SMEs: The case of family and cooperative firms. Technovation, 27(5), 241–253. Cantú,C.,Claessens, S., &Gambacorta,L. (2022). Howdo bankspecific characteristics affect lending? new evidence based on credit registry data from latin america. Journal of Banking & Finance, 135, 105818 Carbo-Valverde, S., Rodriguez-Fernandez, F., & Udell, G. F. (2016). Trade credit, the financial crisis, and SME access to finance. Journal of Money, Credit and Banking, 48(1), 113–143. Carpenter, R. E., & Petersen, B. C. (2002). Is the growth of small firms constrained by internal finance? Review of Economics and Statistics, 84(2), 298–309. Czarnitzki, D., & Hottenrott, H. (2011). R&D investment and financing constraints of small and medium-sized firms. Small Business Economics, 36, 65–83. De Jonghe, O., Dewachter, H., Mulier, K., Ongena, S., & Schepens, G. (2020). Some borrowers are more equal than others: Bank funding shocks and credit reallocation. Review of Finance, 24(1), 1–43. De la Torre, A., Soledad Martinez Peria, M., Schmukler, S. L. (2008). Bank involvement with SMEs: Beyond relationship lending. The World Bank. De Massis, A., Audretsch, D., Uhlaner, L., & Kammerlander, N. (2018). Innovation with Limited Resources: Management Lessons from the German Mittelstand. Journal of Product Innovation Management, 35(1), 125–146. Dell’Ariccia, G., Kadyrzhanova, D., Minoiu, C., & Ratnovski, L. (2021). Bank lending in the knowledge economy. The Review of Financial Studies, 34(10), 5036–5076. Dessaint, O., & Matray, A. (2017). Do managers overreact to salient risks? evidence from hurricane strikes. Journal of Financial Economics, 126(1), 97–121. Deutsche Bundesbank (2018). Bank Lending Survey - Ergebnisse für Deutschland. https://www.bundesbank.de/de/ aufgaben/geldpolitik/volkswirtschaftliche-analysen/-/ bank-lending-survey-fuer-deutschland-602486, (accessed: 12/08/2018). D’Amato, A. (2020). Capital structure, debt maturity, and financial crisis: empirical evidence from SMEs. Small Business Economics,55 (4), 919–941. Eggers, F. (2020). Masters of disasters? Challenges and opportunities for SMEs in times of crisis. Journal of Business Research, 116, 199–208. Escoz Barragan, K., Becker, F. S. R. (2024). Keeping pace with the digital transformation-exploring the digital orientation of smes. Small Business Economics, pp. 1–25. European Central Bank (2018). Published speeches, European Central Bank. https://www.ecb.europa.eu/press/key/date/ 2012/html/index.en.html, (accessed: 06/01/2018). Eurostat (2024). Structural business statistics (SBS). https:// ec.europa.eu/eurostat/statistics-explained/index.php/ Structural_business_statistics_introduced, (accessed: 31/8/2024). Falato, A., Kadyrzhanova, D., Sim, J., & Steri, R. (2022). Rising intangible capital, shrinking debt capacity, and the US corporate savings glut. The Journal of Finance, 77(5), 2799– 2852. Farre-Mensa, J., & Ljungqvist, A. (2016). Do measures of financial constraints measure financial constraints? The Review of Financial Studies, 29(2), 271–308. Faulkender, M., & Petersen, M. A. (2006). Does the source of capital affect capital structure? The Review of Financial Studies, 19(1), 45–79. Ferrando, A., Popov, A., & Udell, G. F. (2019). Do SMEs benefit from unconventional monetary policy and how? Microevidence from the Eurozone. Journal of Money, Credit and Banking, 51(4), 895–928. Freeman, J., Carroll, G. R., Hannan, M. T. (1983). The liability of newness: Age dependence in organizational death rates. American sociological review, pp. 692–710. Freixas, X., Rochet, J.-C. (2008). Microeconomics of Banking. MIT press. Giannetti, C. (2019). Debt specialization and performance of european firms. Journal of Empirical Finance, 53, 257–271. Gourinchas, P.-O., Kalemli-Özcan, S., Penciakova, V., & Sander, N. (2020). Covid-19 and SME failures (Vol. 27877). MA: National Bureau of Economic Research Cambridge. Hadlock, C. J., & Pierce, J. R. (2010). New evidence on measuring financial constraints: Moving beyond the KZ index. The Review of Financial Studies, 23(5), 1909–1940. Heid, F. (2007). The cyclical effects of the Basel II capital requirements.Journal of Banking & Finance, 31(12),3885– 3900. Holmström, B., & Tirole, J. (1997). Financial intermediation, loanable funds, and the real sector. The Quarterly Journal of Economics, 112(3), 663–691. Hu, D., Fang, X., & DiGiovanni, Y. M. (2023). Technological progress, financial constrains, and digital financial inclusion. Small Business Economics, 61(4), 1693–1721. Ifo Institute (2019). Business Climate Index. https://www.ifo.de/ en/ifo-business-climate-index, (accessed: 13/06/2019). Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. Kahle, K. M., & Stulz, R. M. (2013). Access to capital, investment, and the financial crisis. Journal of Financial Economics, 110(2), 280–299. Kaplan, S. N., & Zingales, L. (1997). Do investment-cash flow sensitivities provide useful measures of financing constraints? The Quarterly Journal of Economics, 112(1), 169– 215. Karapanagiotis, P. (2024). The R Package markets : Estimation Methods for Markets in Equilibrium and Disequilibrium. Journal of Statistical Software, 108 (2). Kremp, E., & Sevestre, P. (2013). Did the crisis induce credit rationing for French SMEs? Journal of Banking & Finance, 37(10), 3757–3772. 123 472 Small and vulnerable during crises? Firm size and financing... Kurnia, S., Choudrie, J., Mahbubur, R. M., & Alzougool, B. (2015). E-commerce technology adoption: A Malaysian grocery SME retail sector study. Journal of Business Research, 68(9), 1906–1918. Li, L. (2013). TARP funds distribution and bank loan supply. Journal of Banking & Finance, 37(12), 4777–4792. Love, I., Preve, L. A., & Sarria-Allende, V. (2007). Trade credit and bank credit: Evidence from recent financial crises. Journal of Financial Economics, 83(2), 453–469. Mackowiak, B., & Wiederholt, M. (2009). Optimal sticky prices under rational inattention. American Economic Review, 99(3), 769–803. Maddala,G. S., Nelson, F. D. (1974). Maximum likelihood methods for models of markets in disequilibrium. Econometrica, pp. 1013–1030. Mankiw, N. G., & Reis, R. (2002). Sticky information versus sticky prices: a proposal to replace the new keynesian phillipscurve.The Quarterly Journal of Economics, 117(4), 1295–1328. Meinen, P., Soares, A. C. (2022). Markups and financial shocks. The Economic Journal, pp. 1–29. Mundlak, Y. (1978). On the pooling of time series and cross section data. Econometrica: Journal of the Econometric Society, pp. 69–85. Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187–221. Nicolas, T. (2022). Short-term financial constraints and SMEs’ investment decision: evidence from the working capital channel. Small Business Economics, 58(4), 1885–1914. Nilsen, J. H. (2002). Trade credit and the bank lending channel. Journal of Money, Credit and Banking, pp. 226–253. Ozkan, A., & Ozkan, N. (2004). Corporate cash holdings: An empiricalinvestigationofuk companies.Journal of Banking & Finance, 28(9), 2103–2134. Paravisini, D., Rappoport, V., & Schnabl, P. (2023). Specialization in bank lending: Evidence from exporting firms. The Journal of Finance, 78(4), 2049–2085. Presbitero, A. F., Udell, G. F., & Zazzaro, A. (2014). The home bias and the credit crunch: A regional perspective. Journal of Money, Credit and Banking, 46(s1), 53–85. Rajan, R. G., & Zingales, L. (1998). Financial dependence and growth. The American Economic Review, 88(3), 559–586. Raymond, L., & St-Pierre, J. (2013). Strategic capability configurations for the internationalization of SMEs: A study in equifinality. International Small Business Journal, 31(1), 82–102. Research Data and Service Centre of the Deutsche Bundesbank (2020). Balance Sheets of Non-Financial Corporations (USTAN), 1987-2019 https://doi.org/10.12757/Bbk. Ustan.8719.05.04. Schmidt, R. H., & Krahnen, J. P. (2004). The German Financial System. Oxford University Press. Stiglitz, J. E., & Weiss, A. (1981). Credit Rationing in Markets with Imperfect Information. The American Economic Review, 71(3), 393–410. Taketa, K., & Udell, G. F. (2007). Lending channels and financial shocks: The case of small and medium-sized enterprise trade credit and the Japanese banking crisis. Monetary and Economic Studies, 25(2), 1–44. Tengulov, A. (2019). The value of borrowing diversity: Evidence from the financial crisis of 2007-2009. Available at SSRN 2361401. Udell, G. F. (2020). SME access to finance and the global financial crisis. Journal of Financial Management, Markets and Institutions, 8(01), 2040003. Vermoesen, V., Deloof, M., & Laveren, E. (2013). Long-term debt maturity and financing constraints of SMEs during the global financial crisis. Small Business Economics, 41(2), 433–448. Wai, K. W., Liston-Heyes, C., Liu, W., Liu, G., Cowling, M. (2022). Organizational capabilities and SME exports: The moderating role of external funding intentions and managerial capacity. Small Business Economics, pp. 1–15. Whited, T. M., & Wu, G. (2006). Financial constraints risk. The Review of Financial Studies, 19(2), 531–559. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 123 473