1. Introduction: the conventional wisdom

Financial institutions include lending intermediaries, banks, financial markets, and the legal, customary and state foundations of finance.Footnote ¹ Conventional wisdom has it that entrepreneurs could draw on their savings or borrow from family and friends if needed. The esteemed economic historian Michael Postan (Reference Postan1935: 2–3) wrote:

Similarly, Herbert Heaton (Reference Heaton1937: 4) wrote: ‘external supplies of capital were, however, less important than the personal or family funds which the industrialists scraped together and ventured in the new productive equipment’. Heaton mentioned one case of raising a mortgage, by the famous Matthew Boulton. Otherwise, his account suggests that the entrepreneurs drew on their savings or borrowed largely from family or friends. This became the conventional wisdom. As Alex Trew (Reference Trew2010: 988) put it: ‘The general argument runs that an individual entrepreneur, especially a good one, could find the start-up capital required or use reinvested profits to expand as and when conditions allowed’.

The purpose of this essay is not to overturn this conventional wisdom. Using savings or borrowing from family and friends may have been the norm. But we need a fuller picture and there are some grounds to be sceptical. Thomas Ashton (Reference Ashton1961: 26, 180) expressed doubts: ‘It is true that self-financing was a marked feature of the period, but it would be an error to consider it as universal … The country dealer or manufacturer often needed more capital than could be obtained from his own resources or those of his partners and friends’. Rondo Cameron et al. (Reference Cameron, Crisp, Patrick and Tilly1967: 11, 55) also questioned the ubiquity of self-financing: they pointed out that short-term credit was sometimes used to finance fixed investment in buildings and machinery. Peter Mathias (Reference Mathias1969: 177) noted wryly: ‘As more case histories reveal the number of instances in which this generalization about banks not financing industrial investment was broken, the generalization may itself come under some suspicion’. But still we do not have adequate evidence to dismiss the conventional wisdom.

Especially beyond the trusted circle of family and friends, larger loans often would require collateral, as with a mortgage. Legal and other institutional arrangements for mortgaging evolved only slowly, yet there is some evidence of loans obtained via mortgages. The records of attorneys that have been consulted show substantial mortgaging activity in the 18th century (Anderson, Reference Anderson and Crouzet1969a; Miles, Reference Miles1981). In her study of the financing of the West Riding Woollen Industry, Pat Hudson (Reference Hudson1986: 85) demonstrated that the mortgage market ‘developed rapidly in the eighteenth century through the activities of intermediaries such as attorneys, and much mortgaging and land transfer was occurring by the end of the century’. But Robert C. Allen (Reference Allen1992: 104) noted the lack of estimates of the overall volume of mortgaging. We know relatively little about the extent and nature of borrowing before the 19th century, or for what it was used. A general picture of the role of finance is lacking.

The 17th and 18th centuries saw expanding British trade and colonisation. There were many expensive and risky overseas ventures that required long-term finance. If financial institutions were sufficiently developed to provide this, then it is difficult to understand the claim that they were not also deployed in the aid of entrepreneurial activity in the homeland.

The role of finance in the modern era was underlined by Joseph Schumpeter (Reference Schumpeter1934). Later empirical research underlines its importance. In a macroeconomic study, Peter Rousseau (Reference Rousseau2003) amassed evidence for the Dutch Republic (1640–1794), England (1720–1850), the United States (1790–1850) and Japan (1880–1913). His econometric analysis indicated that, in all cases, the emergence of financial institutions including financial markets were crucial in enabling growth and industrialisation.

Pointing to experiences in many countries, Trew (Reference Trew2010: 988) noted that the conventional wisdom concerning finance in the British Industrial Revolution ‘runs against the large body of empirical evidence … that suggests that financial development is strongly correlated with, and perhaps leads, the level of economic growth’. Multi-national evidence was surveyed by Ross Levine (Reference Levine, Aghion and Durlauf2005: 865) who concluded:

The role of finance is crucial because it no longer limits economic growth to the flows of savings and profits. With finance, entrepreneurs can expand more rapidly. Was Britain an exception? Did it manage to take off industrially without the extensive help of financial institutions?

Britain pioneered industrialisation. Because they faced British competition, the countries that followed were under greater pressure to restructure (Gerschenkron, Reference Gerschenkron1962). Their states were impelled to intervene more extensively in their economies, and to promote the development of banking and financial markets. But this still leaves open the question on whether underdeveloped financial institutions restrained British growth.

There is at least one way of reconciling the conventional wisdom with the evidence on the important role of financial institutions. Borrowing from family and others might have been prevalent partly because of the underdevelopment of financial institutions and their inability to supply adequate loans. Postan (Reference Postan1935: 2–3) hinted at this possibility in the quote above. Philip Cottrell (Reference Cottrell1980: 33) and Joel Mokyr (Reference Mokyr1993: 104–105) speculated that if the imperfections in Britain's capital markets had been reduced earlier, then the Industrial Revolution might have been boosted. Larry Neal (Reference Neal, Floud and McCloskey1994) stressed the interconnection of finance and industry and pointed to the slow development of banking and stock markets. Despite the Financial Revolution of 1660–1720 (Carruthers, Reference Carruthers1996; Dickson, Reference Dickson1967; Hodgson, Reference Hodgson2017; Roseveare, Reference Roseveare1991; Wennerlind, Reference Wennerlind2011), the relatively immature and precarious state of financial institutions and financial markets may have acted as a hindrance on economic development during the 18th century (Hoppit, Reference Hoppit1987; Williamson, Reference Williamson1987).

This may be described as the Schumpeterian hypothesis, because of his emphasis on the role of finance in supporting innovation. Schumpeter (Reference Schumpeter1934: 69, 73) stressed that an innovating entrepreneur ‘must resort to credit … which cannot like an established business be financed by returns from previous production’. This source of finance is not ‘purchasing power which already exists in someone's possession’ but may be ‘supported by securities which are not themselves circulating media’. With appropriate financial institutions, credit based on collateral can create new purchasing power, as if ‘out of nothing’. Schumpeter may have placed too much stress on the strict necessity of finance for innovation. But the hypothesis captures the spirit of his broader argument.

As yet, we have insufficient data to test this hypothesis. This scoping paper provides no new empirical evidence, and it does not resolve the foremost issues of dispute. Instead it frames the argument and points to some key areas and empirical resources where research needs to be done to begin to answer some key questions. Later research is needed to devise strategies for testing the hypotheses involved.

2. The Schumpeterian hypothesis in perspective

The Schumpeterian hypothesis might help to explain that, despite growing innovation and industrialisation, Britain's overall rate of growth from 1760 to 1820 was very slow – in the region of 0.2% per annum in GDP per capita. In the same period, industrial output per capita increased by an average of 0.7% per annum. Many innovations took place from 1760 to 1820, but they did not stimulate growth above 1%. In part this was due to the relatively small size of the innovating sectors in the early years. But it is likely that their growth was constrained. In the following 50 years growth picked up significantly. From 1820 to 1870, GDP per capita growth averaged 1.1% and the growth of industrial output per capita roughly trebled to 2.0%.Footnote ³

Growth was held back before 1820 for several reasons. The Bubble Act of 1720 put severe restriction on the formation of joint stock companies. It was not repealed until 1825. In 1714 the highest legal rate that could be charged for a loan was lowered to 5% and this ceiling lasted until the 1820s, when the most restrictive usury regulations were repealed. The usury laws created disincentives to lend money. Unable to charge high risk premiums on loans, credit was rationed, and lenders were more inclined to ask for commensurate collateral (Temin and Voth, Reference Temin and Voth2013). Progress with land registration was belated and slow, thus inhibiting the use of landed wealth as collateral. (More on that later.) Patenting inventions was cumbersome and expensive.

Another factor was war. Britain was involved in military conflict, with at least one other major power, in 86 of the 127 years from 1689 to 1815 inclusive. Then followed 98 years of Imperial hegemony and Pax Britannica. War has positive and negative economic effects. Military procurement can stimulate innovation, industrial production and employment. For example, a more accurate method of boring an iron cannon was patented in 1774. For Boulton and Watt this innovation was vital for the development of their steam engine (Roll, Reference Roll1930: 25). On the negative side, war exacerbates uncertainty. Some economic historians have argued that state spending and borrowing for war crowded out industrial and other private investment.Footnote ⁴

The spread of financial markets is constrained by communication technology. The English state began to develop a postal network in the 16th century, facilitating the conveyance of documents and valuables (Heblich and Trew, Reference Heblich and Trew2019). Until the 19th century brought the railways and the telegraph, nationwide communication of information was no faster than the speed of a horse.

Peter Temin and Hans-Joachim Voth (Reference Temin and Voth2013) emphasised the negative effects of government restrictions on interest rates, banks and joint stock companies. Adverse government policies impaired financial institutions, which in turn restrained growth. As Temin and Voth (Reference Temin and Voth2013: 185) concluded: ‘The apparent paradox of slow growth in times of rapid technological change disappears when we examine the role of private finance’.

If government had played more of an enabling, rather than a constraining role, then by how much would the rate of development of financial institutions increased? On current evidence it is difficult to say. And were such changes in government policy feasible? They would have required major shifts in ideology and understanding. Although government policies are important – for good or ill – they are not the only causal factors. Britain was ‘paying the penalty for having been thrown into the lead’ (Veblen, Reference Veblen1915: 132). The development of financial institutions depends on detailed knowledge and experimentation, which is often painstakingly slow. Other countries learned from the British experience and developed more rapidly.

Cameron et al. (Reference Cameron, Crisp, Patrick and Tilly1967: 32–35) estimated that the capital invested in English banking (excluding the Bank of England) increased by a factor of 2.2 from 1775 to 1800, and again by a factor of 1.5 from 1800 to 1825. Total banking assets, as a proportion of English national income, increased from 15.2% in 1775, to 27.9% in 1800 and to 29.6% in 1825. These figures show a large rise in the size of the financial sector in the final quarter of the 18th century.

Another rare depiction of the overall size of the British financial services sector appears in Stephen Broadberry et al. (Reference Broadberry, Campbell, Klein, Overton and Leeuwen2015: 164) with estimates at 50-year intervals. According to their data, the volume of financial services was about the same in 1700 than it was in 1650. But from 1700 to 1750 this sector increased by a factor of 2.6. This suggests some growth in financial services before the beginning of the Industrial Revolution. From 1750 to 1800 it tripled by a factor of 3.1. From 1800 to 1850 the increase was again a factor of 2.6. This is consistent with later evidence from other countries that the growth of financial services and markets precedes and enables industrial growth (Levine, Reference Levine, Aghion and Durlauf2005).

More evidence at the microeconomic level may offer a way forward. A geographic microeconomic analysis of data from 1817 to 1881, by Stephan Heblich and Alex Trew (Reference Heblich and Trew2019), found evidence of causal mechanisms through which banks enabled industrial growth. Their study suggests that industrial employment was positively related to the activity of the country banks. Currently, we are unable to test the Schumpeterian hypothesis definitively because too little empirical research has been conducted for the 18th century. But we can summarise existing knowledge and identify needed areas of empirical research.

Note that the Schumpeterian emphasis on financial institutions largely concerns the conditions and enablers of innovation, not its drivers. The hypothesis is compatible, for example, with the stress by Joel Mokyr (Reference Mokyr2009) on the energising role and innovative potential of Enlightenment science and culture. It can also be dovetailed with the emphasis on the roles of ideas and innovations in the study of Deirdre McCloskey (Reference McCloskey2010, Reference McCloskey2016a, Reference McCloskey2016b) as long as her over-hasty dismissal of institutions is annulled.

The Industrial Revolution has been a foremost topic in economic history for well over a century, but our knowledge of 17th- and 18th-century financial institutions and activities is still very patchy. Temin and Voth (Reference Temin and Voth2013: 4) noted that, with some exceptions ‘banks and other forms of intermediation are hardly ever mentioned in modern treatments of the Industrial Revolution’.

The possibility of a ‘shortage of capital’ is considered in the following section. The paper then moves on to consider the evolution of key institutional underpinnings of finance before 1820. There is a particular focus on the growing alienability of land and on the legislation that made mortgaging more practicable.

3. Was there ‘no shortage of capital’?

We need to flag up the ambiguity of the word capital. As several authors have noted, it can mean several different things. Most prominent is the confusion between capital goods and finance capital.Footnote ⁵ Terms like ‘the rate on return on capital’ perpetuate this ambiguity. In this section, we consider whether or not 18th-century industrial entrepreneurs faced a shortage of funds.

Some researchers have looked at interest rates, on the presumption that they are affected by monetary supply and demand (Clark, Reference Clark1996; Pressnell, Reference Pressnell and Pressnell1960; Quinn, Reference Quinn2001). If interest rates were high, then this might indicate a shortage of finance capital. Low rates might indicate a surplus. This exercise has not brought a clear verdict. Sidney Pollard (Reference Pollard1958: 221–222) pointed to the early lack of integration of British financial markets and concluded that interest rates were a poor guide. Stressing the restraints of usury laws, Temin and Voth (Reference Temin and Voth2013) came to a similar conclusion. Fluctuations in interest rates were often related to political crises or wars. Overall, the determinants of monetary supply and demand were unstable, uneven nationwide, and constrained by usury laws and other legislation.

We should also remind ourselves of the economic meaning of demand. It means neither need nor want. As Adam Smith (Reference Sraffa1976 [1776]: 73) put it: ‘A very poor man may be said in some sense to have a demand for a coach and six; he might like to have it; but his demand is not an effectual demand’. Any ‘effectual’ demand for finance must be backed by security or means of future payment. An entrepreneur may be in dire need of additional finance, but unable to express an economic demand for it, because of a lack of secure collateral or of perceived repayment capacity. Collateral was often crucial in creating demand. Any unbacked need for money would have no discernible effect on interest rates.

Nevertheless, several prominent economic historians remained assured that capital was available. François Crouzet (Reference Crouzet, F. and Crouzet1965: 187–188) argued that ‘the eighteenth century capital market seems, to twentieth century eyes, badly organized, but the creators of modern industry do not seem to have suffered too much from its imperfection … English industry … seems to have overflowed with capital’. Accordingly, H. John Habakkuk (Reference Habakkuk1962: 175) argued that if there had been a greater need for finance, then financial institutions would have grown up to satisfy it:

Note that Habakkuk confused need with demand. His claim that financial institutions automatically grow up to meet needs is countered by more recent, extensive evidence from several countries (Emenalo et al., Reference Emenalo, Gagliardi and Hodgson2018; Levine, Reference Levine, Aghion and Durlauf2005; Rousseau, Reference Rousseau2003). An economic demand for finance cannot be expressed without the legal and institutional machinery of property, collateral, pricing and contract. As argued below, this institutional machinery was partly and crucially dependent on a series of legislative acts, which were neither easy nor automatic.

Another approach was to look at the comparative usage of fixed and variable capital goods in the early phases of the Industrial Revolution. For Pollard (Reference Pollard1964: 304), in his seminal study, fixed capital is ‘buildings, machinery, etc.’ and circulating capital is ‘stocks, stores, etc.’. So here by capital he meant capital goods. The argument is that greater deployments of fixed capital goods would require greater tranches of long-term finance. Perhaps the fixed capital goods could act as collateral for the loan. But that is not made clear. Pollard (Reference Pollard1964: 313) concluded: ‘the need for fresh capital, especially for fixed assets, has been less than is often supposed, since much capital could be generated from within the existing domestic manufacturing and associated credit systems’. Fixed capital goods of lower value were taken to imply a lesser need for long-term finance.

Pollard (Reference Pollard1964: 308) concluded that ‘the banks provided little long-term capital because little long-term capital was demanded’. Pollard thus endorsed Postan's (Reference Postan1935) supposition that internal savings or loans from family and friends were the prevalent sources of finance. Neal (Reference Neal, Floud and McCloskey1994: 152) called this the ‘Postan-Pollard story’. But Pollard went further than Postan, by suggesting that savings or loans from family and friends were adequate, as well as the normal means of finance. Similarly, Ronald M. Hartwell (Reference Hartwell1965: 173) claimed that ‘the capital needs of early industrialization were modest’. Similarly, Sushil Khanna (Reference Khanna1978: 1889–1892) asserted that ‘the requirements of fixed investment were very modest and the threshold of entry into factory production quite low … there was no “capital scarcity” in 18th century England’. Noting ‘how low the capital-labour ratio was during the Industrial Revolution’, Khanna (Reference Khanna1978: 1897) then claimed that ‘capital equivalent to about four months' wages was enough to start a firm’. Then in parenthesis: ‘Boulton and Watt, who had the monopoly of steam engines for 25 years, launched their firm in 1775 with a capital of only £3370’.

The celebrated Matthew Boulton and James Watt partnership has been studied in detail (Cule, Reference Cule1935; Roll, Reference Roll1930). Does the fact that their steam engine enterprise was launched with ‘only £3370’ confirm that finance was adequate for their industrial ambitions? No. When they started in 1775, Boulton and Watt did not themselves manufacture many steam engines. Instead, they generally patented their designs and allowed others to manufacture the engines on licence. They drew up and sold plans and acted as consulting engineers. According to Eric Roll (Reference Roll1930: 24–26), the ‘most important’ reason for this delegated arrangement ‘was undoubtedly the lack of sufficient capital’ to finance and set up works themselves. It was not until 1795 that they focused on manufacturing steam engines for sale.

Hence, apart from their Soho premises near Birmingham, Boulton and Watt owned limited fixed capital goods in the early years. But that did not mean that their financial needs were met. The early decades of their business were fraught with financial difficulties. Roll (Reference Roll1930: 38) noted their ‘serious lack of capital which … hampered the growth of the business and more than once presented the partners with the prospect of immediate collapse’. Postan (Reference Postan1935: 4) also wrote on the precarious state of the Boulton–Watt partnership in its early years:

Cule (Reference Cule1935: 28, 83) recorded several mortgages by Boulton – in 1769 raised in Amsterdam with the premises near Birmingham as partial security, another in 1778, and yet another in 1787 or 1788. Addressing other entrepreneurs, Heaton (Reference Heaton1937: 4) noted that ‘fixed capital requirements’ were not necessarily large: ‘Yet they were frequently large enough to harass and perplex those who needed funds for building or equipping a plant of their own; for they were often the last straw on a back that already bore a heavy load’. Heaton added in a footnote: ‘We know of … the strain on Boulton's resources when he was erecting his Soho factory’. Heaton (Reference Heaton1937: 5) summed up Boulton's desperate need for finance. The firm survived ‘after many years of grim abstinence, of pared family budgets, and of frantic efforts to find supplementary funds outside’.

The assumed relationship between fixed capital goods and the need for finance capital seems partly driven by the ambiguity of the word capital. Fixed capital and circulating capital goods are seen to relate to long-term and short-term finance, respectively. In reality, the connection is much looser. The example of Boulton and Watt shows clearly that the ratio of fixed to circulating capital is at best a weak indicator of the need for finance. It also demonstrates that, contrary to Habakkuk (Reference Habakkuk1962: 175), financial institutions did not spring up to rescue entrepreneurs in dire need.

Robert Allen (Reference Allen2009) showed that England (compared with France and elsewhere) had a relatively high-wage economy for some time before the Industrial Revolution. Allen also noted that Britain had abundant cheap sources of energy, particularly coal. He claimed that interest rates provide little indication of a need for finance. Instead, Allen turned to the substitution of capital goods for labour, prompted largely by the lower price of the former relative to the latter.

In his criticism of Allen's argument, Mokyr (Reference Mokyr2009: 268–272) pointed to evidence suggesting that relatively few inventions were designed to save labour. Firms will try to save on all costs, not simply those from the relatively expensive inputs. Above all, innovative effort is always constrained by the extent of useful knowledge.

Thomas Newcomen produced an early steam engine in about 1710. Its main use was to pump water out of mines. Hundreds of these engines were installed in the 18th century. Watt greatly improved the Newcomen engine by adding a separate condenser, more than doubling its fuel efficiency. But Boulton and Watt ‘had little prospect of supplying many of their engines to collieries where coals were cheap and where the saving of fuel was of little importance’ (Roll, Reference Roll1930: 67). The cheapness of coal was here a barrier, not a spur, to technological progress. Boulton and Watt got around this problem by offering their engines to the Cornish tin and copper mines. Unlike the Midlands and North of England, there was no coal in remote Cornwall: it was shipped there from South Wales. The local expensiveness of coal gave the Watt engine an advantage. But also in that county, there was a severe shortage of the skilled labour required to assemble and maintain the engines (Roll, Reference Roll1930: 60–66). Parts for their steam engines had to be transported by road or canal to Bristol, from where they were conveyed by coastal vessels to their Cornish destinations. There, the superior Watt engines quickly replaced the Newcomen machines. This strategy established the Boulton and Watt business before they developed and manufactured rotative steam engines after 1795.

Relative prices of factor inputs clearly mattered. But this story is not about substituting relatively cheap capital for relatively expensive labour. A Cornish tin mining enterprise circa 1780 had the choice of a Newcomen engine, or a slightly smaller, equally powerful, more expensive, and hugely coal-saving engine by Boulton and Watt.Footnote ⁶ There are several kinds of capital goods involved in these options, including coal, and two types of steam engines. Both skilled and unskilled labour are relevant. Neither capital goods nor labour are homogenous. The oversimplified rhetoric of capital obscures this heterogeneity.

We have no way of ascertaining the price of aggregated capital goods without knowing their individual prices. As Piero Sraffa (Reference Smith1960) showed, the value of aggregate capital goods is not independent of relative prices, wage levels and profits. Yet, it is relative prices that are said to drive the choice of combination of capital and labour. The reasoning is circular. The forgotten Cambridge capital theory debates showed the severe limitations of Q = f(K, L) and of the related idea of marginal substitutions between these aggregated factors of production (Harcourt, Reference Harcourt1972). Aggregated prices of ‘capital’ relative to ‘labour’, as displayed by Allen (Reference Allen2009: 138–139), are highly problematic for this reason. As Mokyr (Reference Mokyr, Floud and McCloskey1994: 15) pointed out:

Capital goods are needed for production through time. They can also be used as collateral for borrowing money. Yet, much of the focus of economic historians has been on savings, which come out of profits, as the main options for raising finance. The relative neglect of possible collateralisation of fixed assets is unfortunate. Collateralisation requires developed financial and legal institutions to deal with property and debt.Footnote ⁷

4. The institutional foundations of modern finance

Financial institutions have a very long history. Bank depositories existed in Babylonian temples and in India almost 4,000 years ago. They were also found more than 2,000 years ago in ancient Greece, the Roman Empire and Qin Dynasty China. These ancient banks stored gold and other portable wealth for clients, provided credit and kept records. Their lending function was constrained, partly because non-portable wealth was more difficult to use as collateral. From the 12th century, banks in Italy acted as money changers and provided credit using future agricultural outputs or mercantile profits as security. Fractional reserve banking emerged later, in the Netherlands. It spread to England in the 17th century. Banknotes and cheques were developed, avoiding the need to transport gold or silver (Neal, Reference Neal2015).

Noting the ancient roots of banking, McCloskey (Reference McCloskey2010: 138) concluded: ‘Finance and saving and investment cannot have been crucial’ for modern development, as places such as Florence, Athens or Beijing ‘would have innovated us into the modern world’. This overlooks the primitive state of banking, by modern standards, in ancient China, ancient Greece and even medieval Florence. Neal (Reference Neal2015) saw the beginnings of ‘modern finance’, with stable currencies and long-term public debt, in Genoa in the 15th century. But political fragmentation and wars prevented the full development of a modern financial system in medieval Italy. Finance-driven economies build on both experience and legislation to establish contractual rules, concepts and templates to deal with complex financial obligations. Craig Muldrew (Reference Muldrew1998) explained how these institutions developed in England from the 16th century. Modern banking systems also depend on central banks and markets for debt. Created in 1668, the Swedish Riksbank was the world's first central bank. The Bank of England followed in 1694.

Rather than supporting industry or commerce, the primary reason for the establishment of the Bank of England was to finance state expenditures, particular for war. Nevertheless, Neal (Reference Neal, Floud and McCloskey1994: 180–181) described the Bank of England as helping to develop a ‘web of credit … anchored securely in the City of London’. The number of banks increased. There were about 24 London banks in the 1725, 52 in 1785 and 70 in 1800. In the UK provinces there were about a dozen country banks in 1750, 119 in 1784 and 280 in 1793, reaching to a peak of 707 in 1812. Country banks played an important and growing role, but they were often insecure, and their numbers fluctuated, due to failures and amalgamations.Footnote ⁸

Institutions to establish land as collateral for loans, and markets for debt, were limited in the UK until well into the 18th century. Crucial was the emergence of institutions making debt itself saleable or ‘negotiable’. Limited markets for debt had appeared centuries earlier in China, France, Italy and the Netherlands, but their legal basis was then underdeveloped. Contracts ordinarily involve legal obligations to deliver goods or services in exchange for money. Exchanges of promissory notes involve instead the purchase of a promise, and originally this was not recognised as a valid contract in law: the selling of debt was not sanctioned by legal recognition of the transfer of the obligation to its purchaser. Major legislative changes were necessary to make this possible.

In England in the 17th century the ‘blundering attempts’ (Beutel, Reference Beutel1938: 840) of common law courts to deal with the negotiability of debt led businessmen to press Parliament for robust legislation. Repeated statutory legislation and experiences of case law up to the 1750s were required to consolidate negotiability (Lawrence, Reference Lawrence2002).

Modern finance involves a complex web of contractual obligations. Fractional reserve banking has a cumulative effect on money creation by commercial banks as it expands the money supply beyond the scale of the deposits alone. Any debt is funded by current assets, or by claims owed by a third party. The purchaser of debt receives the right to an asset that itself can be used as collateral to borrow. Credit money thus feeds on itself. But this all depends on a legal structure of enforceability, a fractional reserve system backed by private and state assurances and sufficient confidence that debt can be redeemed. Once legal institutions supporting collateralisable property, credit money and the sale of debt were in place, a new dynamic was unleashed.

But this new system is vulnerable to waves of speculation, causing unsustainable booms and damaging slumps. In his history of central banks in several countries, Charles Goodhart (Reference Goodhart1988) noted their movement from being bankers for the government, to taking responsibility for the whole financial system. Then the central bank no longer treats other banks as rivals but takes some responsibility for their welfare.

Crucially, the Bank of England did not become established as the institutional lender of last resort until about 1760, after a series of financial panics. ‘On the eve of the Industrial Revolution it took its first step towards the assumption of the powers and responsibilities of central banking … the lender of last resort [was] a function of much greater significance than its former role as a fiscal arm of the crown’ (Lovell, Reference Lovell1957: 15–17).

It is reasonable to describe what was consolidated in Britain in the 18th century as capitalism (Hodgson, Reference Hodgson2015a). Contrary to McCloskey (Reference McCloskey2016b: 93) – who again dismisses too speedily – capitalism is not defined as the accumulation of capital goods, or ‘piling brick on brick’ as she put it. It is finance that is definitionally central to capitalism. In part, as Schumpeter (Reference Schumpeter1939: 223) explained, ‘capitalism is that form of private property economy in which innovations are carried out by means of borrowed money, which in general … implies credit creation’. Accordingly, as Geoffrey Ingham (Reference Ingham2008: 73) concluded: ‘Capitalism is distinctive in that it contains a social mechanism by which privately contracted debtor-creditor relations … are routinely monetized’.Footnote ⁹

To consider whether the slow development of the British financial system restrained industrial growth until sometime in the 19th century, we need evidence on the extent of creation of credit through the use of financial institutions in the 18th and 19th centuries. Security or collateral are the foundations of credit. The main store of wealth was land. Among other things, we need evidence on potential and actual mortgaging by entrepreneurs and shareholders.

Economists have paid insufficient attention to the institutional conditions required to mortgage assets or use them as security for loans. For example, in ‘the economics of property rights’ property is regarded as mere possession or control. Legal title is seen as significant only if it aids control of the asset. Otherwise, ‘property’ is simply what you control, and you have a ‘property right’ even if you have stolen it (Alchian, Reference Alchian1965; Barzel, Reference Barzel1989; Furubotn and Pejovich, Reference Furubotn and Pejovich1972). Critics point out that de facto possession does not necessarily constitute a right. This dismembered view of property ignores its multi-faceted legal nature (Cole and Grossman, Reference Cole and Grossman2002; De Soto, Reference De Soto2000; Heinsohn and Steiger, Reference Heinsohn and Steiger2013; Hodgson, Reference Hodgson2015a; Reference Hodgson2015b; Honoré, Reference Honoré and Guest1961; Steiger, Reference Steiger2006; Reference Steiger2008). Possession (usus) is only part of the real story. Ownership is not simply a matter between buyer and seller, but it requires some legal authority. As John R. Commons (Reference Commons1893: 110) put it: ‘in the end, the actual title to property rests on the sovereign power of the state to enforce its decrees’. Consequently, the ‘economics of property rights’ places no emphasis on the possible use of property (including land, buildings and patents) as collateral. The concepts of collateral and mortgage are absent from the studies of Alchian (Reference Alchian1965), Furubotn and Pejovich (Reference Furubotn and Pejovich1972), Barzel (Reference Barzel1989) and others in this genre. But they are vitally important for economic development.

5. Land alienability and mortgageability in England before 1700

If land cannot be sold or its use-rights transferred, then it cannot be used as security for loans. Much English land was bought and sold, at least from the 12th century. Four county-based studies in England from the 13th to the 15th centuries suggest that sales of land were increasingly frequent, among free peasants from the 12th century onwards (Harvey, Reference Harvey1984). Habakkuk (Reference Habakkuk1981: 213) noted that the land market developed during the 16th century, but ‘the general series do not enable one to judge whether more land was sold in, say, the 1720s than in the 1620s, or compare the level of sales in the eighteenth century as a whole with the level in the century before 1640’. Here too have little information to go on.

Although the essays in Harvey's (Reference Harvey1984) book show peasants trading in freehold land growing from the 12th century, there is no mention of any piece of land being mortgaged. Neither ‘mortgage’, ‘collateral’ nor ‘security’ appear in the index of the volume. Studies in the book edited by Chris Briggs and Jaco Zuijderduijn (Reference Briggs and Zuijderduijn2018) confirmed that mortgaging in medieval England was rare. When it occurred it was almost always between geographically proximate lenders and borrowers. Institutional arrangements for mortgaging generally favoured the lenders: the borrowers ran severe risks of losing their land.

Some medieval mortgages involved the transfer of both usufructuary rights and legal title to the lender, for the period of the loan (Briggs and Zuijderduijn, Reference Briggs and Zuijderduijn2018). One reason for this arrangement was to circumvent the usury laws. Before 16th century legislation that made limited interest payments legal, such usufructuary mortgages were among the rare cases of mortgaging.Footnote ¹⁰ Peasant freeholdings changed hands, but much other land was not alienable. There were enduring restrictions on selling landed property, known as entails. Many entails enforced primogeniture, ensuring that a landed estate passed from one generation to another through the eldest son. But even when the courts limited the scope of entails in 1614, these were replaced by voluntary and widespread ‘strict settlements’ that had similar effects, and prevailed until the 19th century (Allen, Reference Allen2012: 65; Beckett, Reference Beckett1984; North et al., Reference North, Wallis and Weingast2009: 89–9).

Entails and strict settlements ‘restricted the uses to which land could be put. … Holders could seldom sell, swap, or mortgage property under their control’ (Bogart and Richardson, Reference Bogart and Richardson2011: 245). Strict settlements stubbornly endured, largely because the wealthy elite wanted and protected them. Owners were disinclined to sell or mortgage buildings or land that had been in their family for generations. Loss of land meant loss of status and privileges.

Much land was set aside as commons. Around 1700, about one-quarter of arable land in England was held as commons, where villagers shared rights to the use of pastures, water sources or woods (Bogart and Richardson, Reference Bogart and Richardson2011: 247). This land could not be sold or mortgaged.

Parliament became involved in the process of making more land alienable. Estate acts undid strict settlements. Statutory authority acts were used to develop infrastructure. Not all enclosures resulted from acts of parliament, but many did. Dan Bogart and Gary Richardson (Reference Bogart and Richardson2011: 249–250) gathered data on the numbers of estate, statutory authority and enclosure acts from 1700 to 1830. They showed that legislative reform of landed property rights was sluggish from 1700 until about 1750 and then took off rapidly, with the strongest growth trends coming from enclosure and statutory authority acts. The annual rate of legislative activity to make land saleable increased about fivefold from 1750 to 1810. Further research into land sales and mortgaging in this period would be helpful to confirm or deny the dramatic picture suggested by the legislative data.

In the 17th century the mortgaging of inheritable copyhold lands became more common. The lord of the manor owned the land but passed lifetime or inheritable usus and usus fructus rights to a copyholding tenant, who paid a nominal rent. If the copyhold were inheritable, then it could be sold on or mortgaged, under the jurisdiction of the manorial court. Hence copyhold transfers were open and under scrutiny, and the manorial court registered the land and the transactions upon it. In contrast, freeholder mortgages lacked any land registry or assembled public record.

Juliet Gayton (Reference Gayton, Briggs and Zuijderduijn2018) found no evidence of a copyhold mortgage before the 17th century. She saw this absence as due to the unreformed usury laws. Using a sample of manorial records, she found a pronounced rise in the number of copyhold mortgages from 1606 to the time of the Restoration. She noted that most borrowers and lenders lived within 15 miles of each other. There was little recorded use of financial intermediaries. In Gayton's sample there was a decline of recorded copyhold mortgages after 1660. This could reflect a shift of mortgage proceedings from the manorial to the common law courts, particularly after the reform of mortgage laws in the 1660s, as discussed below. Other studies do not show a decline in copyhold mortgages (French and Hoyle, Reference French and Hoyle1999).

We lack information on how much land was alienable in the 17th century. But we do know that even if it could be sold, obtaining a mortgage on freehold land was prohibitively risky and expensive before 1670. Allen (Reference Allen1992: 102–103) explained:

The 17th century saw the first steps in developing modern mortgage law. Numerous landowners petitioned the courts for more favourable legislation. But mortgage law ‘remained confused at least into the 1670s’. Allen (Reference Allen1992: 104) continued:

As David Sugarman and Ronnie Warrington (Reference Sugarman, Warrington, Brewer and Staves1995) pointed out, there is some irony in the fact that privileged landowners lobbied so hard for a change in mortgage law that would boost their hold over their landed estates, and yet it would eventually lead to a release of finance capital that would help the rise of the rival industrial bourgeoisie to power.

Empirical research by Grajzl and Murrell (Reference Grajzl and Murrell2021a, Reference Grajzl and Murrell2021b) into English case law shows that the rules of mortgaging became under greater court dispute from about 1640 and deliberation upon them rose until 1690. Subsequently, disputes remained at around that level, to begin to fall after about 1720. This indicates that mortgage law became substantially settled before the middle of the 18th century.

From the 17th century, forms of mortgaging evolved in attempts to get around the legal encumbrances tied up with landed property. One option was to obtain a mortgage on leasehold rights, rather than on the legal title to the property, as collateral (Anderson, Reference Anderson and Crouzet1969a; Clark, Reference Clark1998: 60; Neal, Reference Neal, Floud and McCloskey1994: 163). But such transactions were still fraught with difficulties. Eventually, as the legal encumbrances to alienation diminished, mortgaging based on legal title as security became more prominent.

Data on land sales gathered by Clark (Reference Clark1998: 74–76) cover the period from 1500 to 1910 but show no instance of a mortgage before 1680. Allen (Reference Allen1992: 104) wrote: ‘The volume of finance raised by mortgage grew rapidly in the late seventeenth century’. He referred to William Petty (Reference Piketty1927: vol. 1: 247) who had estimated in about 1660 that capital raised with land as security equalled only 2.5% of the value of English real estate. Allen further noted that in the mid-1690s Charles Davenant had estimated that loans secured on land amounted to £20 million, which equalled about 8% of the value of English and Welsh property. Allen regretted the lack of later estimates.

In 1690 the economist Nicholas Barbon helped to develop the National Land Bank, which would mortgage real estate. But, opposed by the Treasury and Parliament, this scheme foundered in 1697 (Clapham, Reference Clapham1966: 33–34; Rubini, Reference Rubini1970). It took several decades to build up financial institutions under which alienable land could be readily mortgaged. Much land in the 1690s could not yet be used as collateral. Further changes in land ownership and land registration were necessary, and these were slow in coming.

Contrary to Douglass North and Barry Weingast (Reference North and Weingast1989) and Daron Acemoglu et al. (Reference Acemoglu, Johnson, Robinson, Aghion and Durlauf2005), the key change in property rights due to the Glorious Revolution of 1688 was not an improvement in their security. In England, landed property was relatively secure from the 12th century (McCloskey, Reference McCloskey2010; Fukuyama, Reference Fukuyama2011: 418–420; Hoppit, Reference Hoppit2011; Ogilvie and Carus, Reference Ogilvie, Carus, Aghion and Durlauf2014). Increased rights of alienability (normally a precondition of mortgageability) imply changes in the nature of landed property rights. Evidence suggests that possibilities for alienability increased, noticeably after 1750 and into the 19th century (Hodgson, Reference Hodgson2017). Property was relatively secure. What was needed was increased opportunities for land alienability combined with other institutional arrangements to facilitate the use of land to obtain credit.

6. Raising mortgages in the 18th century

As Benito Arruñada (Reference Arruñada2012) explained, there need to be safeguards to ensure that the mortgagor has genuine title to the land, and its boundaries are not under dispute. A partial solution is the formation of land registries, as Hernando de Soto (Reference De Soto2000) has pursued in contemporary Peru. But as Arruñada (Reference Arruñada2017) further elaborated, an operable system of land registration requires trustworthy financial intermediaries plus legal and administrative institutions that are efficient and relatively free of corruption. Even today, only a minority of countries fulfil these conditions (Dam, Reference Dam2006).

By 1663 a land registry was in operation called the Bedford Level, covering 703 square miles including the whole of the Isle of Ely and adjacent areas with Huntingdonshire, Norfolk and Suffolk. By 1700 there was some clamour for further land registries, to facilitate land mortgaging and the raising of capital. Land registries, with records of conveyances and mortgages, were established in the West Riding of Yorkshire in 1704, the East Riding of Yorkshire in 1707, Middlesex in 1708 and the North Riding of Yorkshire in 1735. Remarkably, these records seem rarely explored by economic historians.Footnote ¹² By comparison, the Low Countries had a more extensive and much better organised system of land registration as early as the 17th century (Van Bochove et al., Reference Van Bochove, Deneweth and Zuijderduijn2015). In the UK, local systems of land registration proceeded very slowly and took centuries to develop. Progress was patchy. But from the end of the 17th century there is even some evidence of mortgaging in parts of England that lacked land registries.

Mortgages did not always finance industrial investment or entrepreneurship. Much land was mortgaged to pay for conspicuous consumption, buildings, marriage settlements and idle pleasures (Cannadine, Reference Cannadine1977, Reference Cannadine1980; Joslin, Reference Joslin and Pressnell1960: 176).

In the 18th century the Bank of England had little to do with mortgages (Joslin, Reference Joslin1954: 175–177). An Act of Parliament of 1708 prevented private banks with more than six partners from issuing bank notes. Some private banks developed a limited but growing mortgage business. In their study of Hoare's Bank in London, Temin and Voth (Reference Temin and Voth2013: 50) reported: ‘Mortgages were the single most important security offered in the years before 1710’. Joslin (Reference Joslin and Pressnell1960) also noted several cases of lending via mortgages by the London private banks in 1739–1784. But again the evidence is limited. What is certain is that all UK private banks remained small until after the Joint-Stock Bank Act of 1826, which replaced the 1708 Act and heralded a more effective banking system.

In his study of country banking, Leslie Pressnell (Reference Pressnell1958: 344) noted the growing need for long-term finance, including for major agricultural improvements: ‘Loans by way of mortgage were procurable in London, from banks and insurance companies and through the multitudinous lawyers of the capital’. Pressnell (Reference Pressnell and Pressnell1960: 187) concluded: ‘Much the most important means of raising money privately in the eighteenth century was the property mortgage’. Mathias (Reference Mathias1969: 175–176) cited several enterprises that obtain loans from banks, but without detailing whether there was any mortgage or collateral. His list includes the Duke of Bridgewater, who opened the Lancashire canal in his name in 1761, plus Richard Arkwright and other industrial entrepreneurs. The evidence we have so far is very patchy, but it reveals cases where mortgages were raised on land or other assets to finance industry or infrastructure.

Infrastructural development was crucial to prepare the ground for industrialisation. In her study of the English turnpike trusts, Brenda Buchanan (Reference Buchanan1986: 230) noted the early development ‘of the concept of the “mortgage” beyond its landed origins. … Before 1793 the tolls formed the only security, and these were assigned to three landed trustees. Each mortgagee had “all the Right Title and Interest in and to the said Tolls”’. Ward (Reference Ward1974) found a similar innovation in mortgaging arrangements in the early canals and river improvements, where the rights to the barge tolls where sometimes mortgaged. He noted a mortgage on a landed estate for river improvements as early as 1698. His data also show that the Sun Insurance Company provided some mortgages.

To whatever use the loans were put, mortgage activity was significant. By using the records of attorneys, Anderson (Reference Anderson and Crouzet1969a, Reference Anderson1969b) established that mortgaging occurred in Lancashire throughout the 18th century, and Miles (Reference Miles1981) and Hudson (Reference Hudson1986) found significant evidence post-1750 of borrowing and mortgaging in the West Riding of Yorkshire. Anderson (Reference Anderson1969b: 18) noted examples of mortgages on urban dwellings in Lancashire: ‘it is clear that the practice of borrowing and lending on mortgage had taken root among all classes right from the beginning of the [eighteenth] century’. Mortgaging became no longer the exclusive device of the rich.

The impressive study by Philip Hoffman et al. (Reference Hoffman, Postel-Vinay and Rosenthal2019) of peer-to-peer lending from 1740 to 1899 in France, serves to warn us of the amount of ‘dark matter’ that may be out of sight, in Britain and elsewhere. Hoffman and his colleagues showed that peer-to-peer lending in France was huge, in rural as well as urban areas, almost century before French industrialisation and the rise of its banks. In 1770 about one-third of French families borrowed money. Much of this lending involved mortgages and was facilitated by the use of notaires. This borrowing was not generally from family and friends. It could no rely on trust alone. As well as acting in their capacity as state officials to authorise and record contracts, the notaires often served as financial intermediaries, by matching lenders with borrowers and arranging loans. Before 1850, lenders and borrowers were rarely more than a day on horse apart, because the lender might have to visit the borrower to retrieve the debt. Nevertheless, lending flourished.

Hoffman and his colleagues were able to access data on the extent of this lending because of surviving central records. The closest equivalent positions in Britain are the attorneys or scriveners. But these have never been state officials and no central British records of such lending exist. Finding comparable data of peer-to-peer lending in Britain will be a much harder task. But the French case suggests that much British financial activity, while legal, may hidden in the archives.

This raises the question if finance was available in 1740, why did France not industrialise until well into the 19th century? The answers are beyond the scope of this paper, but may lie in pre-1789 state despotism alongside a fragmented legal system, revolutionary turbulence from 1789, the slow development of larger private companies, or the underdevelopment of wider financial markets, including markets for debt (Carruthers, Reference Carruthers1996: 23).

In Britain, when peer-to-peer and bank lending were inadequate, borrowers approached other financial institutions. John (Reference John1953: 155) highlighted the role of insurance companies in providing mortgages: ‘In the first half of the [eighteenth] century the only office to make great use of mortgages was the Sun Fire, to be followed later by the Equitable’. The mortgaging function of insurance companies expanded after 1750. Mingay (Reference Mingay1963: 37) noted: ‘By the end of the eighteenth century the Equitable and Sun Fire Offices alone had some £776,000 invested in mortgages’.

Further archival research might help to resolve some important questions. For instance, it is clear that as the Industrial Revolution progressed, many joint stock enterprises were financed by shareholders. But where did the shareholders obtain the money to buy their shares? It would be interesting to see how many shareholders mortgaged land or buildings to finance their investments.

More land could be mortgaged as enclosures of common lands progressed. Pressnell (Reference Pressnell1958: 350) wrote:

Much of this evidence has not been disinterred, despite the analysis of enclosures having been a hot topic for economic historians. Mingay (Reference Mingay1963: 97–98) and Liam Brunt (Reference Brunt2006: 87) noted the option of mortgaging enclosed land. More sceptically, Michael Turner (Reference Turner1981: 236–237) pointed out that the extensive use of enclosed land to raise funds by mortgage ‘has yet to be demonstrated’ and ‘yet to be investigated’. Turner suggested that Pressnell may have overestimated the likelihood of this happening. But we simply do not know, because adequate research has not been done.

In summary, it seems that that there was substantial mortgaging activity by end of the 18th century. But mortgage risks were high. For example, as Ellis Powell (Reference Powell1915: 118, 124) noted: ‘The lending of money on farm mortgages for agricultural improvements was the principal case of the failure of 240 country banks in 1814, 1815, and 1816’. In the crisis of 1825–1826, 60 country banks failed (Cameron et al., Reference Cameron, Crisp, Patrick and Tilly1967: 27). The emergencies of war and successive banking failures led to reforms and restructuring of the banking system, including the Country Bankers Act of 1826. Such observations are consistent with the Schumpeterian hypothesis that the rickety and constrained state of British financial institutions was one of the factors that held back growth during much of the Industrial Revolution.