key: cord-0050318-gkd1wb09 authors: Collier, Paul; Mayer, Colin title: Reforming the UK financial system to promote regional development in post-COVID Britain date: 2020-08-29 journal: nan DOI: 10.1093/oxrep/graa028 sha: 075eaab112cdb6f9d983f028437b493858ff8dcb doc_id: 50318 cord_uid: gkd1wb09 The UK government faces a massive post-COVID problem in restructuring failing companies and rebuilding its already depressed regions. A missing part of the solution is to link government as well as private-sector funding to the financing of small and medium-sized enterprises (SMEs) in the regions. The institutional structure that is required has precedents in the UK, which can be used as the basis for reforming the funding of its SMEs. This paper is about the promotion of small and medium-sized enterprises (SMEs) post-COVID in the provinces of the UK. We provide evidence on why they are disproportionately important during times of high uncertainty such as the present and argue that public finance is required to pump-prime the local private institutions that are needed to fund them. There are three propositions that underpin the paper. The first is that to offset job loss from shifts in demand patterns, market opportunities will need to be met by speculative private investment. Yet the rise in uncertainty due to COVID will discourage such ventures. The aversion to uncertainty is probably greater in large firms with formal processes for approving decisions, than in entrepreneurial SMEs driven by 'animal spirits'. But while such SMEs may be willing to invest, British finance is skewed against them. The provision of external investment finance for SMEs from banks depends upon collateral, while venture capital, which finances on prospects, depends upon the commercial viability of generating the extensive local knowledge on which judgments must be based. The second proposition is that due to the extreme concentration of British finance in London, the SME disadvantage in finance is much greater in the provinces: there is a 'liability of distance' which reduces both bank lending and venture capital. This has already contributed to the extreme spatial differences in productivity: high productivity does not diffuse from firms in the south-east to provincial SMEs: they are 'hubs without spokes'. Hence, if the finance for provincial SME investment can be increased, it can be expected both to speed the offset of job losses and help the longer-term process of regional rebalancing of productivity. The final proposition is that neither the private sector nor government can provide such finance acting alone. The private sector has not done it even pre-COVID, presumably due to high set-up costs of the necessary intermediary institutions and limited initial opportunities in poorer regions. Regional prospects can only be transformed by a sustained range of public actions such as provision of infrastructure and training which lie beyond the scope of this paper. But a valuable additional component would be to establish locally based risk-capital for SMEs around the country, and this is what the present proposal aims to achieve. We begin by describing the investment-SME problem in section II. We then set out the regional-social problem in section III. In section IV, we turn to a solution: combining the exceptional capacity of national government to raise finance, with incentivecompatible devolved decision-taking to venture capital with the capacity to deploy it to the minority of SMEs with genuine opportunities for scale-up. Section V concludes the article. COVID is both reducing aggregate demand and is widely expected to change its pattern. Spanish evidence, based on changes in consumer spending from credit card data, finds that during the first weeks of the lockdown, overall spending fell by 49 per cent, but within this were large variations: the average for the ten best-performing categories was an increase of 150 per cent (Carvalho et al., 2020) . Currently, this presents firms with two investment uncertainties: whether to incur the costs of maintaining their enterprises and retaining staff so that they are ready to ramp production back up if and when demand revives, and whether to incur the costs of meeting expanded demand which may prove to be temporary. During the recovery phase some of these uncertainties will be resolved: many firms will decide that their markets have permanently declined and so shed workers. At that point, continuing wage subsidies to discourage labour-shedding become ineffective and impede necessary resource reallocation. However, for these job losses not to cause aggregate long-term unemployment, they will need to be offset by new opportunities being identified and taken to scale through investment. While COVID disruption has reduced the return on the stock of capital, the shift in opportunities will have raised the return on such scale-up investment. Whether, once aggregate demand has been restored, these new opportunities are capable of re-employing the workforce productively, cannot yet be known. Our proposition is that addressing the current lack of finance for investment by provincial SMEs is an instance of 'no regret' actions that-even under conditions of radical uncertainty-are worth doing regardless of subsequent knowledge. It is a necessary, though by no means sufficient condition for restoring employment, a point to which we return in section III. Britain's SMEs are already disadvantaged relative to larger companies in access to investment finance. This general proposition is not controversial: 'it is widely accepted that the UK lacks sources of finance for growing innovative firms looking to scale up' (Mason, 2020) . But while COVID has created opportunities, its consequences have also increased the uncertainty facing investment decisions. Our proposition is that this uncertainty differentially disadvantages finance for SMEs. We know from a new global analysis of large firms, based on 12,000 public CEO briefings to shareholders, that by May 2020 'uncertainty' was the word they most commonly used to describe the problems posed by COVID (Hassan et al., 2020) ). This applies even to the relatively straightforward decision of whether to invest in maintaining productive capacity for a post-COVID recovery in demand, but more forcefully to the decision of whether to finance new ventures that are hopeful and innovative. Whereas aversion to uncertainty (sometimes referred to as 'ambiguity aversion') is known to be widespread, until recently it was assumed to be a 'bias' which affected only unsophisticated actors and would consequently not affect large firms which have many opportunities to diversify risks. However, recent work that has tested this presumption has found it to be precisely wrong: sophisticated decision-takers are more uncertainty-averse than others (Hong et al., 2019) . We do not yet know why this is the case, but a plausible explanation is that the formal process of board approval required in publicly quoted companies is designed to produce challenges to any irreversible commitment of resources that cannot be quantitatively substantiated. Yet this inability is intrinsic to uncertainty, since the frequency distribution of outcomes cannot be known (Manski, 2013; Kay and King, 2020 ). An implication is that although large firms continue to have access to finance for investment during COVID uncertainty, they may be reluctant to commit resources to an expansion of production. Indeed, since the risks of employing labour have unambiguously increased, the investments most readily sanctioned by the board of a large firm will be the risk-reducing substitution of capital for labour: unlike employees, robots can neither catch a virus nor sue for having caught it. Such investment will, however, compound the employment problem. Three distinctive features of SMEs make them important in these circumstances. Being omnipresent across sectors and locations, they are better placed to discover new market opportunities, and they may react to them more quickly, being less encumbered by the bureaucratic decision structures inherent to large firms. Further, facing a higher cost of capital than large firms, expansion is likely to be more labour-intensive. And central to our subsequent discussion, since SMEs are usually being managed by founderowner entrepreneurs, they may be motivated to invest by 'animal spirits' rather than quantitative calculations. Since under uncertainty the anticipated returns on investment cannot be reduced to probabilities, the lack of quantitative evidence must be substituted by some other form of knowledge if resources are to be committed. The theory of the epistemological basis for action in the face of uncertainty has recently been formalized by psychologists and sets out a detailed specification of a substitute termed a 'conviction narrative' (Tuckett and Nikolic, 2017) . The concept is intricate, but for present purposes, the only aspect that need concern us is that it is a form of narrative and that narratives are often critical in decision-taking. They are critical because many decisions taken by individuals are based primarily on knowledge learnt from a networked peer group rather than on knowledge directly learnt through individual observation. Such knowledge is stored and communicated through narratives that circulate within the group, such as the entrepreneurs in a city. Evolutionary biology has shown that this style of decision-taking predominates because it is evolutionarily efficient (Henrich, 2017) . The formal procedures of large firms are designed to protect decisions from the influence of unquantified narrative propositions. But SME decisions are taken by individual entrepreneurs and so are more likely to be influenced by them. Narratives provide two sources of reassurance that can enable action despite uncertainty. Even if all the SME entrepreneurs in a networked group are basing their view on the same limited information, consensus among peers provides the illusion of common knowledge. This is well-understood to provide some spurious reassurance that can justify commitment (Banerjee, 1992) . Further, knowledge held in common facilitates coordination within the group (Thomas et al., 2014) . This offers some betterfounded confidence since many investment decisions are interdependent and so confer externalities on each other (Minniti, 2005; Venables, 2018) . Hence, a confident narrative about new opportunities can enhance the mutual prospects of successful investment. Narratives also often entangle causal propositions with emotions and so generate 'irrational exuberance'. Since many founder-owners of SMEs passionately believe in their firm, this entanglement is perhaps analogous to what Keynes meant by 'animal spirits'. Hence, SME entrepreneurs, having less sophisticated procedures than large firms for taking investment decisions, may be less deterred by uncertainty from wanting to invest in new opportunities. However, an SME entrepreneur may wish to try a perceived opportunity yet be unable to do so. Many such opportunities can only be properly tested if implemented at some threshold scale: below this, production costs are prohibitively high so that the market potential remains unknown, as in Akerlof and Holden (2016) . This threshold requires finance, whether from the own resources of the entrepreneur, or from wealthholders and their intermediaries. They anticipate that many opportunities will require external finance from venture capital, and that due to information asymmetries the market for it will not be competitive: financiers will capture rents. If external finance is needed, this pushes the decision problem under uncertainty one stage back from the willing-but-finance-constrained entrepreneur, to the wealth-holder. Wealth invested in an SME is radically less liquid than owning the equity or bonds of a publicly quoted company, and monitoring the firm is more costly per unit of investment. Individual wealth-holders can reduce these risks and costs by using intermediaries: banks and venture capital funds. To raise funds, the intermediary will need to convince wealth-holders that it can distinguish between entrepreneurial exuberance and genuine opportunities. This task is demanding: Anyadike-Danes et al. (2009) tracked a cohort of business start-ups and found that half of the increase in jobs came from only 6 per cent of them. Banks are able to lend to SMEs on collateral, but are not equipped to lend on their assessment of an investment prospect, and indeed are not permitted to do so by financial regulators. Identifying these rare genuine scale-up opportunities is the role of venture capital intermediaries, vital both for commercial viability and employment generation. Since SME investments are illiquid, reassurance depends upon a capacity for intensive collection of information prior to investment, with the continuous and proximate monitoring needed to generate firm-specific knowledge until the investment matures, and active involvement in management during which the intermediary can mentor the company. This is complex, so intensive in tacit knowledge (Nightingale, 2009) . In mentoring, the venture capitalist combines the codified knowledge learnt from past training in finance and management practice, with this tacit knowledge learnt from experience in SME scale-up. A good venture capitalist devotes substantial time to acquiring and conveying this knowledge to entrepreneurs and investors (Mason, 2020) . Even with this intensive process, a venture capital intermediary expects to make many mistakes, but through its efforts it develops a track record which converts an initial state of radical uncertainty into quantifiable risks. But the tasks impose considerable fixed costs, creating a threshold scale of operation problem for the intermediary. The market may not merely be uncompetitive: if proximity is necessary, in some places no intermediary may be viable and so risk-capital to SMEs will be unavailable: for example, there is no venture capital firm in Wales, a region of 3m people. The uncertainty arising from COVID consequently affects both the desire to invest and the ability to finance it differently for large firms and SMEs. In respect of the desire to invest, as the boards of large firms react to the increase in uncertainty, investment projects will be less likely to be approved. In contrast, the disruptive demand shifts caused by COVID create new opportunities that SME entrepreneurs may seize with enthusiasm. In respect of access to investment finance, large firms almost always have access because their equity and bonds are readily marketable. SMEs are only able to raise finance for investment if they have collateral or access to venture capital. COVID weakens SME access to banks because it creates uncertainty as to the valuation of assets such as houses and stock, and, as we argue in section III, many SMEs may have no access to venture capital. This yields a testable hypothesis about the consequences of uncertainty. In large firms, what drops is not their access to finance but their appetite for it. In contrast, in SMEs, what drops is not the appetite but their access. The Global Financial Crisis of 2008 caused a sudden leap in uncertainty equivalent to COVID. Besley et al. (2020) investigate its effect on access to credit by firm type. They find precisely the pattern of access that we predict: SME creditworthiness drops more than for large firms and takes much longer to recover. Britain is now the most spatially unequal major society in the OECD (McCann, 2018) . Since 1980 growth has been heavily skewed towards London and the south-east, where labour productivity is now 70 per cent above that in the rest of the country. The skew in the spatial allocation of public expenditure on infrastructure and research funding has contributed to this divergence. For example, each pound of private-sector spending on research and development is matched by a pound of public spending in London, but only 25 pence in the West Midlands (Jones and Forth, 2020) . Social expenditures and outcomes have become similarly divergent. Education spending and attainments are considerably higher in London than elsewhere, as are health spending and outcomes (GBD Associates, 2018). As this has become politically salient, regional economic convergence, and especially the need to raise productivity in the provinces, is a now major policy priority. Localized productivity arises from agglomerations that create mutual benefits to firms and workers: common location confers external benefits on others that are proximate (Venables, 2018) . The theory of agglomeration predicts that the spatial allocation of activity is ex ante indeterminate but ex post locally stable. All clusters enhance productivity by harnessing cluster scale-economies, but clusters of knowledge-intensive firms are particularly valuable since they provide complementary opportunities for a workforce to further raise productivity by the high return consequent upon investment in pertinent skills. The interdependency of spatially-specific investment decisions extends well beyond firms. Investment in skills matters, and these decisions are taken by households envisaging futures for their youth, and training colleges planning courses. Developers decide on investments in housing and speculative commercial property; local governments decide on infrastructure; central government decides on research funding for universities; and universities decide whether to use it in ways that engage local business. Each of these has the potential to enhance the productivity of the others, but each is a distinct community, and so will be liable to base decisions on its own narratives. For the equilibrium of a poor region to be reset, these interdependent decisions will need to change. There have been many previous Whitehall policy interventions to redress regional imbalance, but all have failed. We do not claim that the lack of investment finance for SMEs in poor regions is either the sole cause or its provision the sole solution to this formidable problem. Indeed, given the above interdependencies, it seems likely that a package of complementary policies will be required. But poor regions will need many more productive jobs and most of them will need to come from SMEs. Yet to date, the geographic distribution of British SME scale-ups has been heavily skewed towards the south-east (Mason, 2020) . Haldane (2018) shows that to an internationally unusual extent, technology and best practice have not diffused spatially across SMEs: our innovator firms, evidently mainly in the south-east, are 'hubs without spokes '. McCann (2020) notes that this cannot be because of the labour market, since Britain has unusually high inter-regional labour mobility. Our argument is that one plausible explanation for the lack of spatial diffusion among SMEs is spatially differential access to finance. Typically, the most important source of bank finance for SME investment is a loan secured on the value of the owner's house. Here, the extreme divergence in house prices between the south-east and provincial Britain has created major differences in access (Henley, 2020) . Similarly, we know that job creation by an SME is positively associated with a financial inheritance, which is consistent with finance being a constraint, and reinforcing the importance of the skew in house prices. Yet rather than this skew in bank collateral being mitigated by venture capital, it has been reinforced. The spatial distribution of venture capital finance has been highly skewed to the south-east, which, in the most recent 3 years for which data are available, has received between 61 and 77 per cent of it (Mason, 2020) . Why might this be? Clearly one possible explanation is that, given the spatial skew in productive agglomeration, the south-east is where the opportunities are concentrated. This must, to an extent, be correct and indeed contributes to our argument. The smaller size and reduced opportunities for scale-up in poor provincial cities make venture capital commercially unviable and once this happens, access for genuine opportunities is lost. The exceptional lack of diffusion noted by Haldane suggests that there must indeed be many opportunities for productivity catch-up for provincial SMEs that are being missed. It would be implausible to explain this by a lack of entrepreneurial talent or animal spirits: historically such talent has been widely diffused and until 1980 the provinces were growing more rapidly than London. Our explanation, which builds on the larger agglomeration problem, is that the spatial skew in the allocation of finance has been compounded by the extreme, and internationally exceptional, spatial concentration of Britain's finance industry in London. Given the need of venture capital for localized tacit knowledge, this has created a 'liability of distance' problem in access to finance (Lee and Brown, 2017) . It is thus consequential both for slowing the pace of post-COVID recovery and impeding regional convergence. In a market economy, public policy intervenes to correct divergences between social and private benefits and costs. In Britain, these have recently altered due to both political change and COVID. In respect of benefits, the social value of generating a new productive job in a poor region has risen considerably. This is partly because of the new awareness of the extent of regional divergence, and partly because COVID has created concerns about how to offset national job-loss through SME scale-up: on recent experience, the prospects of such scale-ups look to be much worse in provincial Britain than the south-east. In respect of costs, there are two important changes. One concerns the cost of finance, which has fallen for government, while becoming less available for provincial SMEs so that its implicit price has risen. The other concerns the cost of reaching the rare instances in which provincial SMEs have genuine opportunities for scale-up. The financial intermediaries currently available to the government are the banks. But as discussed above, they lack the knowledge necessary to identify such opportunities. Hence, the only option currently available is for the government to fund all SMEs that want to borrow by guaranteeing bank loans, and letting the banks subsequently foreclose on the many that, due to the structural shifts generated by COVID, will not be able to repay. Understandably, the banks are reluctant to play this role, which would damage their image, and so they are currently advocating a public 'bad bank' to which they can pass such loans. The skills to identify SME opportunities, to the extent that they exist in Britain, are in the venture capital industry, but in many provincial cities such an industry has been commercially unviable: private costs have exceeded private benefits. COVID and levelling up have between them created three wide gaps between social and private costs and benefits and that this constitutes an opportunity for public policy. The first is between the high cost of finance to a provincial SME with a scale-up opportunity, and the low cost of finance to the government. The second gap is between the cost to the venture capitalist of imperfectly identifying and supporting such opportunities and the vastly higher cost to government of doing so through indiscriminate support via the banking sector. The third gap is between the benefit to government from supporting such SME opportunities as could potentially be supported by venture capital, and the lower benefit to the venture capitalist of doing so. These three gaps constitute an opportunity for public-private partnership that blends public purposes with private capacities. Two types of new financial institution that would work together are needed: one national, the other local. At the national level what is needed is a public investment/development bank. It could initially hold the claims on the companies currently funded through government COVID loan schemes, but also manage further public resources generated by government borrowing. This would be used to support public purposes such as provincial job creation and zero net emissions, through the acquisition of equity and loans (Besley and Stern, 2020) ). Our proposal is that this new national bank would allocate some of this finance to a second class of institution: locally based venture capital teams, which would be commercial but subject to the discipline of being compliant with stated public purposes. While venture capital firms are currently missing in many provincial cities, the opportunity arises because they could now be started at a scale of funding that could overcome the threshold problem, and with a low cost of public capital that would make even modest returns profitable. The national development bank would channel public finance into them, and itself raise further finance, typically from large Impact Investment funds. Similarly, the local venture capital enterprises could supplement this public finance with money from wealth-holders raised on commercial terms, and which might be particularly attractive to the rapidly growing number of Impact Investors. Both the staff of SME venture capital firms, and the Impact-Investing wealth-holders who back them, are typically motivated both by an intrinsic purpose to address social problems and the extrinsic financial benefit they derive from successful selection of SMEs: public and private purposes, while not coincident, are not incompatible. Venture capital would not be expected to bail out all SMEs, but rather to identify those that are commercially viable and separate them from those that need to be put into insolvency, perhaps by passing them to a 'bad bank'. This second category of institution is not such a radical innovation as it might seem. They would be related to what existed in the UK until 1990 in the form of 3i, Europe's most successful venture capital firm, and its precursor, the Industrial and Commercial Finance Corporation (ICFC). ICFC was owned by the UK clearing banks and the Bank of England and focused on the financing of small manufacturing companies in their early stage of development (Coopey and Clarke, 1995) . It often took equity stakes in firms, but it did not have direct representation on their boards. It undertook active screening and monitoring of borrowers and, unusually for British clearing banks, its loan officers had a high degree of technical competence and commitment to long-term lending. It developed a degree of industrial expertise that meant that it could bridge the gap between finance and business which emerged when the British banking system lost its local roots in the nineteenth century. ICFC was a considerable success. Following losses in its first 3 years to 1948, it made substantial profits thereafter. The number of its investments went up by a factor of ten between 1954 and 1984 and investments by ICFC were regarded by companies as a mark of quality certification. After it was permitted by its bank shareholders in 1959 to raise external funds, ICFC expanded to become the largest provider of growth capital for unquoted companies in the UK. In 1973 ICFC acquired Finance Corporation for Industry (FFI), a sister company also formed in 1945, and in 1983 the combined group was renamed Investors in Industry-3i. At that stage, its investments became focused on venture capital, initially on start-ups, early-stage and development capital and, by the end of the 1980s, it became by far the largest provider of venture capital in the UK. The banks and the Bank of England sold their stakes in the 3i Group in 1987 and in 1994 the company was floated on the London stock market with a market capitalization of £1.5 billion. Thereafter 3i shifted from being a funder of SMEs and a main provider of venture capital in Britain to a private equity firm. Its repositioning contributed to the switch in UK venture capital from early stage in the mid-1980s to management buyouts by the end of the 1990s. Like the nationalized industries ICFC had a public purpose, but it avoided their fate by decentralized and disciplined pluralism. It was decentralized through delegating authority to local branches, whose staff were knowledgeable about business as well as finance and were expected to develop a real understanding of their client businesses. It built disciplined pluralism through its shared ownership between the commercial banks and the Bank of England, protecting it from short-term pressures of markets and politics, and it remained true to its purpose of achieving commercial success from growing with its borrowers, rather than profiting from them. What is required going forward are several independent organizations of this second type generating some competition between them to enable benchmarking and reducing the rent-capture highlighted by Akerlof and Holden (2016) . As a major investor in these institutions, the national development bank should play an active role in ensuring that their purposes and values align with the objective of promoting long-term, patient, risk-taking investment in regional-based SMEs. Such national 3i-type organizations in which the power of decision is decentralized to local branches could operate in parallel with independent commercial venture capital on a level playing field since the latter would be eligible for public funds and be subject to the same clearly specified purposes, monitored by the national development bank. This type of organizational design of competition within common purposes is increasingly being used in the corporate context (Mayer and Roche, 2020) . In this paper we have proposed a new strategy for the provision of risk-capital to provincial SMEs that would help both to offset the job losses likely from COVID, and to reduce the current extreme regional economic imbalance. SME investment is distinctive in motivation, constraints, and social value, and its distinctive motivation is potentially helpful during COVID. Most SMEs are run by founder-entrepreneurs whose 'animal spirits' can potentially overcome the uncertainty-aversion that inhibits investment within the more formalized decision processes of larger firms. Since, per unit of investment, they are also likely to be more labour-intensive, they are also distinctively valuable for social purposes. But the distinctive constraints on investment finance facing SMEs are an impediment, and provincial SMEs are particularly disadvantaged. They are often in areas of low house prices and distant from the headquarters of financial institutions which therefore lack local tacit knowledge needed for venture capital. Our proposed solution has been to deploy central government funding into devolved private risk-bearing capital that is invested for a public purpose compatible with, though not identical to, the private interest of a venture capitalist. The result is to catalyse scale-up by provincial SMEs more efficiently than through bank guarantees. Movers and Shakers Measuring Business Growth: High Growth Firms and their Contribution to Employment in the UK A Simple Model of Herd Behavior The Aggregate Consequences of Default Risk: Evidence from Firm-level Data Tracking the COVID-19 Crisis with High-Resolution Transaction Data 3i: Fifty Years Investing in Industry Measuring Performance on the Healthcare Access and Quality Index for 195 Countries and Territories and Selected Subnational Locations: A Systematic Analysis from the Global Burden of Disease Study The UK's Productivity Problem: Hub No Spokes Firm-level Exposure to Epidemic Diseases: Covid-19, Sars, and H1N1 The Secret of Our Success: How Culture is Driving Human Evolution, Domesticating our Species, and Making Us Smarter You Need to Recognize Ambiguity to Avoid it The Role of Innovation in the UK's COVID-19 Recovery Radical Uncertainty: Decision-making for an Unknowable Future Innovation, SMEs, and the Liability of Distance: the Demand and Supply of Bank Funding in UK Peripheral Regions Perceptions of Regional Inequality and the Geography of Discontent: Insights from the UK Productivity Perspectives: Observations from the UK and the International Arena Public Policy in an Uncertain World: Analysis and Decisions Productivity and the UK's Deficiency in Scale-Ups Putting Purpose into Practice: The Economics of Mutuality Information and Network Externalities Tacit Knowledge and Engineering Design The Psychology of Coordination and Common Knowledge The Role of Conviction and Narrative in Decision-making under Radical Uncertainty Globalization and Urban Polarization