Introduction
For over three decades, from roughly the 1990s to the 2010s, financial inclusion and microfinance—the provision of small-scale financial services to meet the needs of low-income individuals—have been key pillars of development policy.Footnote 1 With the perceived successes of microfinance comes the challenges of economic sustainability, social outreach, and socioeconomic impact. For all the attention microfinance has garnered, much about it is still unknown.Footnote 2 Most current research on microfinance focuses on its impact on the well-being of borrowers and their communitiesFootnote 3 and on the design of credit institutions,Footnote 4 not on the sources of funding or access to capital. These remain unopened black boxes.Footnote 5 This presents a critical problem for the industry because many microfinance providers, initially funded and organizationally reliant on philanthropy, are unable to transition to independent economic viability when they cannot mobilize loanable funds.
Microfinance institutions have two distinct functions. Their primary function is to improve the lives of their borrowers. Their secondary function is to provide financial services to the poor or underserved in a commercially sustainable manner.Footnote 6 These functions are often at odds, as achieving one can undermine the other. Institutional logics—the beliefs that motivate and the practices that achieve the extra-legal, social, and economic goals that determine the behaviors of business organizations—can provide a framework to understand this tension and how it affects the decisions and ultimate success of microfinance institutions.Footnote 7 Although the institutional logics framework has been criticized as being “intuitively attractive, but arguably difficult to define and even harder to apply in an analytically useful manner,”Footnote 8 researchers have applied it effectively in both contemporary and historical contexts in the study of microfinance institutions. For example, by using the institutional logics paradigm, Battilana and Dorado illustrate the challenges faced by financial institutions with conflicting and competing banking (commercial) and development (charity) institutional logics. These organizations struggle to strike the difficult balance between the banking and development logics required to be sustainable.Footnote 9 Funders of microfinance institutions, such as governments or commercial bodies, encounter similar challenges when deciding whether to provide capital, for what purposes, and to whom. Cobb, Wry, and Zhao use the institutional logics framework to assess these bodies’ decision-making processes and their outcomes.Footnote 10 In a historical context, Haveman and Rao, and Haveman, Rao, and Paruchuri, use the framework to study thrift institutions and to show how organizational changes and institutional logics coevolved in the early twentieth century in response to changes in thrifts’ social, economic, and political environments. These institutions shared motivations and beliefs that could be influenced, and these influences could translate into practices that were then shaped by heterogenous local environmental factors and by both political and economic uncertainties. The shift from dissolving thrifts (e.g., individuals in communities coming together to help each other) to permanent thrifts (e.g., saving banks) was influenced by changing socioeconomic trends, such as social and union movements and increased migration that eroded community cohesiveness.Footnote 11
Institutional logics are influenced by local factors specific to both time and place. The long history of microfinance institutions provides current researchers with the ability to better understand the particular influences and effects on these institutions, and thus how to improve the design of policies and structures that will ensure the success of microfinance institutions today. Consider, for example, the financial inclusion of the working classes via savings and credit institutions. This is as critical an issue today as it was in the nineteenth century.Footnote 12 The accessibility of microfinance in the nineteenth century across the countries is shown in Table 1, and it reveals a striking peculiarity: Ireland, despite its relative poverty, had a larger number of formal microfinance providers in the 1830s and early 1840s than the United Kingdom, of which Ireland was a constituent member. Using an institutional logics-informed analytical framework, this study seeks to understand the locally specific factors that led to this anomaly and what they tell us about microfinance today.
Start year and number of savings banks in select countries, per capita
Note: LFS is Loan Fund Societies. TSB is Trustee Savings Banks.
# From 1801 to 1920, Ireland was a constituent member of the United Kingdom of Great Britain and Ireland, so figures from column 2 are included in column 4.
* This date refers to the Dublin Musical Society, the first formal organization that continued to operate for over one hundred years. A loan fund was operated by Dean Swift in the 1720s, but it ceased with his passing.
^ The formal dates of trustee savings banks in Ireland vary, with institutional histories dating banks to 1815 for Belfast, but formal documents refer to 1817.
∆ Is an approximation based on evidence from contemporaneous reports.
Sources: 1820: data inferred from British Parliamentary Papers (BPP), Report from the Select Committee on the Employment of the Poor in Ireland, 18. 1830: data inferred from BPP, Return from Clerks of Peace in Ireland of Transcripts of Rules and Regulations of Loan Funds. 1840–1880: data from various years of Annual reports of the Commissioners of the Loan Fund Board of Ireland.
Col. 2: 1820–1840: data from Pratt, History of Savings Banks in England, Wales, and Ireland; Pratt, Summary of the Savings Banks in England, Scotland, Wales, and Ireland; BPP, Return from each savings bank in United Kingdom, xxviii. 1860–1880: data from Thom’s Directory. Cols. 3-8: data from Wadhwani. “The Institutional Foundations of Personal Finance,” Table 1.
The specific object of this study is the expansion of Loan Fund Societies (LFSs) in the decade before the Great Irish Famine (1845–1849). LFSs were legally established with a conflictual structure in that they were intended to be commercially viable charitable institutions that both provided credit to the deserving and dedicated their “profits” to supporting the destitute. Using insights from surviving qualitative and quantitative data, this study assesses the factors (by conducting process tracing, which emphasizes the “collection of evidence to trace the unfolding of some event or process”) that influenced LFS activities in Ireland.Footnote 13 A decade before the famine, the introduction of a new Poor Law (in 1838) provided keen motivation for the establishment of LFSs. By using LFS profit distribution data, the institutional logics of LFSs are revealed in terms of the implicit factors that influenced their decision-making processes; and by using a discrete choice econometric model, the institutional logics intuition is corroborated. This study finds the following: while difficult local economic conditions should have disincentivized LFS owners from entering businesses or induced them to shut existing ones, incentives to do the opposite were stronger. These included positive economic incentives such as the potential for lower poor rates (i.e., land taxes as discussed below) and higher returns on investments—and social incentives such as an abiding concern for the industrious and destitute poor. These incentives thus motivated LFS owners to enter or expand their businesses. However, operational tensions between the economic and social motivations became salient with legislative changes in LFS business parameters in 1843, and many FLSs, especially those with a narrow market focus, failed. These tensions and challenges are reflected in the choices contemporary microfinance institutions make today in response to changes in the regulation of their industries.
The Microfinance Ecosystem in Pre-famine Ireland
In the late 1700s, the poorer sections of Irish society had little access to formal financial services. They could, however, borrow high interest credit from pawnbrokers, second-hand shops, local shopkeepers, merchants, and farmers. These informal borrowing services were augmented or replaced by LFSs, which were formed in the early eighteenth century,Footnote 14 and by Reproductive Loan Funds (RLFs; discussed below). Trustee Saving Banks (TSBs) provided saving but not credit services. Together, these microfinance institutions provided formal financial services to the poor. The need for such institutions was demonstrable both for the borrowers—the so-called deserving or industrious poor—and for the beneficiaries—the indigent, often disabled, and elderly. Yet LFSs remained relatively rare until the early 1840s because the informal system was seen as adequate, even if not ideal. Then, between 1838 and 1845, a burgeoning ecosystem erupted with more than 550 Loan Fund Societies, which were funded by philanthropic or interest-earning debentures. That is, between these years, 392 LFSs were registered with the Loan Fund Board (LFB), and 161 more LFSs were registered with the London Relief Committee (LRC). Further initiatives traced their lineage to the Reproductive Loan Funds or were developed independently, like the Derry Loan Fund.Footnote 15 These LFSs joined with the 78 government-subsidized TSBs to serve the “industrious poor” and provide sustenance for the indigent. Figure 1 shows the map of these three types of institutions.
Map of Irish LFS, RLFs, and TSBs, c. 1842.
Source: Shapefile from Gregory and Ell, “Irish Poor Law Union and Barony Boundaries.”
New insights into LFS behavior can be garnered from a contemporaneous survey of loan fund societies conducted by H. J. Porter, a land agent, an LFS advocate and a loan fund operator in County Armagh.Footnote 16 Porter highlighted how all counties in Ireland were eventually served by either LFSs or RLFs (e.g., Kerry and Sligo Counties were served only by RLFs at the time of Porter’s survey). For those who could not access bank credit but relied on the very expensive credit provided by pawnbrokers, “gombeen men,” and “meal mongers,” the accessibility and outreach of these institutions was huge. Small deposits were not relied on as a source of capital as LFSs paid interest only on deposits over £5.
More broadly, these charitable LFSs, contemporaneously recognized as a critical component of the Irish financial system,Footnote 17 were “peculiar to Ireland.”Footnote 18 They were seen as unwelcome competition by joint stock banks because of their high rates of interest (which was set by statute and payable on debentures), their ability to lend within the Bank of Ireland monopoly zone, and their promotion by the government as sources of needed finance for an underserved and exploited segment of the economy. A comparison with the commercial banking sector helps to fully appreciate the accessibility and outreach of the LFSs. In 1836 there were 11 commercial banking companies that operated a combined network of 122 branches.Footnote 19 These branches tended to be located in major towns and cities. LFSs were, in contrast, dispersed throughout the island, often in villages and hamlets, and were easily accessible to their clientele. Only one commercial bank, the Agricultural and Commercial (A&C) Bank, made a short-lived attempt to serve the “humbler classes.” It was founded in 1834 and suspended payments in November 1836. The A&C had explicitly aimed to combine the services of a loan fund society and a savings bank. However, even at its peak in October 1836, there were only 44 branches of the A&C in operation, whereas there were ten times the number of LFSs.Footnote 20
Reproductive Loan Funds
The RLFs were formed by an accident of generosity, or “accidental logics.” The circumstances under which the RLFs were “chosen” by the London Relief Committee were contingent upon context, and there was limited control over this context; thus, “what comes first (even if it was in some sense ‘accidental’) conditions what comes later.”Footnote 21 While famines in rural Ireland were a fact of life, a harsher-than-usual famine in the west of Ireland in 1822 led philanthropists in London to raise money to relieve those suffering distress.Footnote 22 Although the subscription was intended only for the immediate relief of the famine, the fund was oversubscribed, thus leaving a substantial surplus.Footnote 23 The trustees of the fund, the LRC, decided to use the surplus to establish income-generating and educational activities in the areas affected. It dedicated about half of the surplus (£40,000) to fund loans to the poor specifically for the manufacture of flax and wool, as the LRC deemed these activities to be a route out of chronic poverty. To design the RLF system, William Hyett, the founder of RLFs in Ireland on behalf of the Irish Relief Committee, visited existing, generally urban, LFSs in Dublin, Limerick, and Ennis that made loans to weavers and other urban laborers.Footnote 24 Mirroring the LFS structure, RLFs—of which there were 161 reported in 1839Footnote 25—worked via local, voluntary, unremunerated trustees.Footnote 26 These trustees disbursed the loans to the poor in the designated areas,Footnote 27 and they reported their activities to the central authority in London. The loans were initially nonmonetary, often instead being items such as looms or seeds with a maximum value of £10 over a twelve-month period. Repayments could also be nonmonetary in nature.Footnote 28 RLFs operated only in Connaught and Munster (excluding Waterford) Counties, the area for which the initial fund was designated.
Loan Fund Societies
Loan Fund Societies, founded before and around the 1836 Loan Societies Act, could trace their ancestry to the philanthropic lending of Dean Jonathan Swift, of St. Patrick’s Cathedral, in Dublin, and of the Dublin Music Society. LFSs were strongly aligned with philanthropy.Footnote 29 Elites, the wealthy, and generally the Church of Ireland clergy played key roles in founding LFSs. Management of LFSs was by wealthy debenture holders—members of the LFSFootnote 30—who volunteered to serve as trustees, provide the working capital, and hire clerks to carry out the day-to-day operations.Footnote 31 Unlike the mutual saving banks of continental Europe, membership was not a requirement for borrowing.Footnote 32 Rather, members “joined” LFSs to provide loans to the industrious poor to enable them to become economically self-sufficient, and thereby better able to weather a normal economic cycle; and to provide sustenance for those—the elderly and disabled— who were unable to care for themselves. While their motivations were noble, they earned a healthy return on their investment. LFSs operated under a legal nondistributive constraint: net earnings could not be redistributed directly to members. This is distinct from Hansmann’s definition of mutual savings banks, in which depositors are not members or owners, and control lies with a self-perpetuating board. In LFSs, members elected the board and borrowers had no voting rights, although technically the latter had the ultimate claim to residual profits.Footnote 33 LFS members did, however, exercise control over the day-to-day running of the society and could, given statutory constraints, say whether net earnings were retained, added to capital, or expended for particular operational or charitable purposes. The LFSs’ charitable character kept members focused on serving the public rather than on their own pecuniary interests.Footnote 34
In contrast to the RLFs, LFSs’ “capital” was comprised of a mix of donations, retained earnings, and, significantly, large deposits, generally debentures. The composition of capital is unknown pre-1845, but Porter suggests that deposits over £50 were both the “most numerous” and “most valuable” in terms of establishing an LFS.Footnote 35 When the Loan Fund Board, the regulator of LFSs, began reporting deposits, 90 percent of LFS capital comprised interest-bearing deposits.Footnote 36 While LFSs paid interest on deposits, as set and required by statute (Table 2), as mentioned above, they restricted this to deposits exceeding £5. Porter noted that working classes were unlikely to deposit in LFSs and advocated, without effect, that lower interest be paid on amounts from £0.25 to £4 to encourage saving.Footnote 37 Small depositors were not the focus of LFS charity because LFSs were not savings banks. Instead, depositors were unlikely to be poor, but were, according to Porter, “comfortable farmers”Footnote 38 or former usurers.Footnote 39 The 1850 report of the Cashel Loan Fund offers a representative example: the members held deposits ranging from £50 to £150.Footnote 40
Business parameters of LFSs
Note: RLF is Reproductive Loan Funds. LFB is Loan Fund Board.
† 6 pence in the pound (there were 240 pence in a pound).
Source: Authors’ compilation.
Unlike other UK charitable institutions, LFSs operated as individual units rather than as branches.Footnote 41 This structure allowed LFS founders to determine the establishment and the location of their LFS. The LFS depositors, locals of some means, were few in number within each district, were usually members of the local Poor Law Union, and could easily be coordinated to provide funding and management services. In 1842, there was a total of 5,867 depositors in all LFSs, which averages to only 19 depositors per LFS. Ensuring borrowers did not borrow from Peter to pay Paul was more difficult. To achieve this, societies requested that the central Loan Fund Board, established in 1836, restrict new LFSs to currently unserved areas.Footnote 42 This position was heartily endorsed by the Loan Fund Board.Footnote 43
While LFS structures were specific, oversight was weak enough to allow for a variety of behaviors that contradicted the Loan Fund Act’s required philanthropic behaviors. Deviations from the act ranged from LFS clerks embezzling funds to small grocers and pub owners setting up shop behind the counter to pseudo-banks establishing themselves under the guise of the Loan Fund Act. These various LFS forms did not deviate from the restrictions on loan size or chargeable interest, and while they may have differed on late fees, the experience for the borrower in respect to the narrow credit transaction was largely similar. They did, however, deviate on motive. This was done explicitly, as in the case of the shop and pub owners who sought not philanthropy but tied sales (of liquor and dry goods, thereby expanding the nature of the transaction) and profits, thus barring the destitute from benefiting. Or it was done deviously, as in the case of the fraud and theft perpetuated by some LFS clerks, thereby reducing or negating the possibility of repeat borrowing because of lack of funds.
The full range of possible motivations for establishing a LFS is summarized in Table 3, in which we have identified and defined four different models of institutional logics. The Loan Fund Acts allowed a number of different organizational motivations to coexist. Anecdotally, there were reports of all types of fund societies listed; however, it is difficult to precisely categorize funds activities according to macro models. By using indicators from the LFB reports, it is possible to gauge the macro models adopted by LFSs in given years.Footnote 44 The majority of registered societies can be classified as either Type 1 or Type 2 LFSs, as shown in Table 3. (For the distribution of the various indicator variables, see Tables A1.1–A1.5, and for the number of LFSs above or below the means of these distributions, see Figures A1.1–A1.4).Footnote 45 Types 3 and 4 were fewer in number, and while they may have adhered to the letter of the law, they also generally ignored its spirit.Footnote 46 This is also reflected in the low number of societies “closed due to irregularities of management” after inspection by the LFB. Between 1842 and 1845, only fourteen LFSs were closed by the LFB, although a caveat is that the LFB could only close an LFS that violated its own rules, and the resources available to the LFB were inadequate to inspect and audit all LFSs on a regular basis.
Institutional logics of RLFs and LFSs
Source: Authors’ compilation.
RLF or LFS?
The empirical distinctions between RLFs and LFSs are important, yet the RLFs’ sizable legal, geographical, and operational distinctions have been understudied.Footnote 47
These distinctions had real and demonstrable implications in terms of regulation and supervision by the LFB, even though their loan structures and interest payables were indistinguishable by the late 1830s. LFSs were required to register with the central Loan Fund Board (located in Dublin), to submit to annual returns, and to undergo annual audits. RLFs, in contrast, although governed by the same act, were exempt from these regulations and instead submitted returns to the London Relief Committee, the provider of their funding. The empirical significance of this is trifold. First, the scale of all RLF loan fund activity was much greater than is generally currently acknowledged. Second, a large un/undersupervised loan fund sector operated on the fringes of the supervised sector, creating the possibility of positive or negative spillovers on LFS formation and lending decisions. Third, the specific initial motivations for establishing the RLFs were the extreme suffering caused by the 1822 famine in the west of Ireland and the perceived need for economic development to lessen (if not eradicate) local poverty. These are distinct from the much more general motivations of LFSs founded over a decade later, which were to make the industrious poor self-sufficient and provide for the indigent. RLFs faced quarrels among their philanthropic funders as to how the problems of poverty in Ireland could be cured. Initially, the universal idea was that peasants’ lives could be improved by the discipline of repaying their loans, which was in line with later discussions on loan funds, such as by C. A. J. Piesse, who was the LFB’s secretary. However, evangelicals argued that the only way to solve Irish poverty was to rid the country of Catholicism, which coincides with their increased proselytizing activities in the 1830s. The disagreements within the charitable society led to accusations of misappropriation by RLF trustees starting in the late 1820s.
The 1838 Poor Law and the Rise of Microfinance
The 1830s was a period of great political turbulence that generated a sequence of significant political and social reforms.Footnote 48 The explosion of LFSs in Ireland coincides with the development of a nascent welfare system financed by a new land tax in 1838: the poor rate. The introduction of the poor rate was an important and unwelcome fiscal shock to Irish property owners, who were already subject to other land-based taxes. For example, there were county fees to pay for local infrastructure, as well as a new tax to replace the tithes supporting the clergy of the Protestant Church of Ireland in predominantly Catholic rural Ireland.Footnote 49
Parliamentary inquiries in the early 1830s investigated the scale of poverty in Ireland. The Whately Commission (1833–1836) advised that some measure of poor relief was required although, specifically, not along the lines of workhouses. The wide-ranging developmental recommendations of the commission were never fully implemented. Instead, the government introduced a Poor Law based on the “new” (1834 variant) English Poor Law.Footnote 50 The introduction of the poor rates—a tax on owners (landlords) and holders (farmers/owners of industrial sites in urban locations) of landed property, payable by “every occupier of rateable hereditaments” (i.e., property that could be inherited)Footnote 51—created a massive fiscal shock to property owners. Ignoring vigorous opposition from both local (i.e., ratepayers) and national (i.e., unionist and nationalist) politicians alike,Footnote 52 the Poor Law system was introduced in 1838.Footnote 53 Given that the majority of the Irish population was rural, the Poor Law was, for all intents and purposes, a tax on rural property owners. And this system did not provide a consistent level of relief for the destitute poor. Rather, it was characterized by unequal provision linked to the distribution of wealth in the many newly created individual Poor Law Unions (PLUs). The boundaries of these unions were designated by taking the market town where the poor-house was to be located and then assigning the surrounding landed estates, ideally, to a single union.Footnote 54
The Poor Law was intended to provide a safety net for the truly destitute— those unable to care for themselves from their own resources. Since contemporaries feared the moral hazard associated with the “comfort” of outdoor relief (i.e., free food rations), the system was designed to discourage (ab)use by requiring entry into a poor-house—of which there was one per PLU—to receive assistance. The stigma associated with this feature made it unpopular with those it was designed to help. Moreover, those applying for relief could be turned away if the poor-house was full. Unlike in England, there was no statutory right to relief in Ireland.Footnote 55 Even with these draconian features, opposition and resentment of the Poor Law by ratepayers remained, leading to attempts to limit their tax liabilities by encouraging various forms of self-help, including savings banks and friendly societies.Footnote 56 The Poor Law and the growth of “self-help” were also related in England. Falling poor relief expenditures led to an increase in self-help activity, and contemporaries attributed “much of the growth in friendly society membership and working-class saving to the reform of the Poor Law.”Footnote 57 Ratepayers had two tools to achieve limiting their taxes: civil disobedience (that is, refusal to pay their poor rates) and establishing LFSs (that is, reduce the target population of destitute poor). They used both. For example, with the former, in 1843 alone, twenty-one unions violently resisted the collection of rates, leading to the deployment of both military and police to enforce the payment.Footnote 58
The Whately Commission’s report and recommendations proposed that “there shall be a loan fund established in each district [Poor Law Union], and that it be administered according to such regulations as the Commissioners shall approve.”Footnote 59 This was specifically to reduce the need for poor relief by bettering the economic resiliency of the industrious poor. The 1836 Loan Society Act occurred after the release of the Whately Commission’s report (and the LFS bill, which was published in June 1836 and enacted in August 1836, also was after the release of the report, which was leaked in April and published in June).Footnote 60 The subsequent LFS legislation in 1836 and 1838 actively encouraged the formation of LFSs to this end. The 1836 legislation purposely aimed to “to afford encouragement to the formation of other institutions of a like kind” by allowing LFSs to charge high interest rates on loans and pay high rates on deposits, the first act to explicitly permit holding of deposits. So successful were these inducements to form LFSs that in 1843 these incentives were significantly lessened by reducing allowable rates (see Table 2 ).Footnote 61
The 1836 act also aimed to “give protection to the funds of such institution” by establishing a governing body, the Loan Fund Board, for “the general control and superintendence of all loan fund societies established in Ireland under the authority of this act.”Footnote 62 The LFB had quite vast supervisory powers “to inspect the books, accounts, and papers of or belonging to such societies.”Footnote 63 LFSs were required to register with and submit a copy of their rules to the LFB. A failure to register or a violation of any rule caused the LFS to be excluded from the benefits of the acts, such as being able to sue for any outstanding loan payments. LFSs (not RLFs, as they had a special derogation) were required to submit annual reports to the LFB, which in turn produced annual reports on the LFS system for Parliament. These are now the main source of data on LFS activity.Footnote 64
Resentment against the Poor Law can be seen in the works of contemporaneous pamphleteers, who argued that LFS formation would decrease Poor Law expenditures, and thereby decrease the pressure on rate payers. Testing whether the presence of LFSs reduced Poor Law expenditures was complicated by the Great Irish Famine, which occurred shortly after implementation of the Poor Law. P. B. Ryan, who was manager of the Thurles branch of the Agricultural and Commercial Bank, opposed the Poor Law on the grounds that it “impos[ed] burdens on the more meritorious classes, in order to support the less worthy.” Instead, Ryan proposed that the profits of a private LFS system that lent to the industrious poor could be used to fully finance a workhouse system for the destitute poor and the old.Footnote 65 Matthew Barrington, another influential pamphleteer, referenced the Poor Law to argue that a public poor relief system could be financed from the profits of microfinance. Barrington suggested that these fund societies would support medical charities as well as “go far in preventing the necessity of Poor Laws, by supporting the aged and infirm, and affording employment to a large portion of the labouring population of the country.”Footnote 66 His attempt to do this in his home city of Limerick received a positive reception. The Dublin University Magazine stated that, although it was a local venture, “we observe in it the beginning of a complete and noble revolution in the system whereby our charitable establishments are now supported, and we regard the success of the sagacious experiment with an interest proportioned to the grandeur of the results which are likely to flow from it.”Footnote 67 The Barrington model, in which profits from lending to the poor supported the poor, was attractive as compared to the idea of supporting the poor by taxing the upper and middle strata of society.
Other pamphleteers saw microfinance as a complement to the Poor Law rather than as an outright substitute. For example, the LFB held in high esteem the opinion of W. L. Shuldham, a land agent in Cork who operated an LFS in Dunmanway.Footnote 68 Shuldham argued that since “the half-employed labouring classes” would be the potential users of poor relief, “surely any plan that held out a rational prospect of diminishing the numbers of people requiring such aid must deserve at least inquiry.”Footnote 69 He was more nuanced in his prognosis and warned against his “zealous friends” who saw LFSs as a substitute for poor relief. Instead, he advocated LFSs as complementary to a poor-law system.Footnote 70 This view was seconded by the LFB and its secretary, the above-mentioned Piesse, who was also an inspector of an LFS.Footnote 71 He stressed the “duty” of the upper and middle classes in supporting LFSs, because “if they do not contribute this aid in enabling the poorer classes to support themselves and their families out of the Poor-House, they must support them in [the Poor-House].”Footnote 72 Ideally, an LFS would be the sole provider of credit in an area within a Poor Law Union, so tax relief could be captured through lower locally assessed poor rates, highlighting the interconnectedness of microfinance, philanthropy, and the Poor Law. In this light, LFSs were seen as complementary agents of contemporaneous social policy.
At the local level, LFS managerial reports made explicit reference to the Poor Law. The manager of the Carrick-on-Shannon LFS wrote that “several industrious tradesmen and poor dealers are most materially benefited. . . . [They] are not ashamed to say that ‘if the loan fund was closed, they should take refuge in the poor-house.’”Footnote 73 The manager of the Duleek LFS asserted “that in very few instances have persons within the sphere of its operations been driven to the necessity of having recourse to the workhouse of the union.”Footnote 74 The manager of the Tobercurry LFS argued that the LFS “has been instrumental in checking mendicancy and has saved many families from the Poor-House.”Footnote 75
Disentangling the exact mechanisms influencing LFS formation is complicated by the interrelation of key motivations: Was the Poor Law itself a direct influence on the formation of LFSs; or was it in anticipation of them, given that bills for the law were published in February, May, and December 1837?Footnote 76 Or was the Poor Law indirectly influential via pamphleteers writing in response to it and arguing for LFSs as a means to reduce the burden of or to replace the Poor Law (given that pamphlets published between 1837 and 1841 explicitly referred to loan funds)? Or were the significant returns promised on debentures issued by LFSs, as set out in the Loan Societies Act as a complementary proposal to aid poor relief, the key motivators for LFS formation in PLUs?
The crux of the argument is that the timing of the implementation of the Poor Law was a predominant influence on LFS formation. The various motivating factors influencing their formation can be gauged from the timing of in a quasi-observational study (Figures 2 and 3). There was a significant increase in the number of LFSs registered with the LFB (itself established in 1836) between 1838 (when there were 50) and 1842 (at the highest peak, when there were 300). This increase in LFSs is somewhat surprising as the formation of the LFB made market entry more difficult and expensive than necessary. A nascent LFS had to register with the LFB; develop and promulgate rules consistent with the Loan Fund Act; and follow the recommendations of various LFB circulars concerning, among other things, proper bookkeeping, formal bonding of clerks and other employees of the fund societies, and performing due diligence on all loan applications (including into the ability of the sureties to make good on a loan should the borrower default). Moreover, an LFS’s accounts were no longer strictly private affairs but audited by the LFB. In the quasi-observational event study, the timing of the increase in LFSs suggests a direct correlation with the introduction of the Poor Law that promoted LFS formation. Further evidence comes from Porter’s survey of LFSs, in which he asked about the year of formation.Footnote 77 Of the 215 surveyed LFSs, 163 provided this information; 61 percent of these 163 societies were formed after 1838. Porter’s findings are corroborated by inferring the year of formation from the LFB reports. These suggest that each of the 300 LFSs in 1842 had been operating, on average, for approximately only four years.Footnote 78 The exceptions to this were the older societies, most notably the Dublin Musical Society, which had operated since the 1700s,Footnote 79 although in 1842 it was barely active.
Timeline of LFB and LFSs, 1829–1845.
Source: Authors’ compilation.
Number of registered LFSs, 1838–1845.
Source: Authors’ compilation.
Revealing Preferences
Profitable Nonprofit
The founders of LFSs—local elites—were strongly influenced by the introduction of the Poor Law. The LFSs could only be of benefit to the poor, and thus to the elites, if they were accessible and provided a beneficial service. LFSs provided small loans (under £10) for short periods (twenty weeks) at low a cost (6d/£, reduced later to 4d/£). Because of these legislated parameters (see Table 2), there was no, or very limited, price competition among loan funds. All loan rates were high, which were made still higher by fines charged for late repayment and the need to compensate sureties, but these costs were low in comparison to the alternatives discussed earlier.Footnote 80 LFS lending rates were relatively cheap, given that contemporaneous pawnbrokers charged as high as 50 APR for loans under £1 and 25 APR for loans between £1 and £10.Footnote 81 Private moneylenders’ rates varied from 25 percent to 100 percent APR.Footnote 82
The £10 cap on loan fund loans coincided with a similar loan ceiling imposed on private pawnbroking.Footnote 83 The average LFS loan size was in the region of £3 to £4 in the 1830s and 1840s (Table 4). The average RLF loan was £2 to £3. The LFS and RLF loans, generally, were made to small traders and agriculturalists (i.e., farmers and/or landless laborers).Footnote 84 These were significant expenses, given their average weekly wage was approximately four shillings six pence (£0.23) in the 1840s; and the “poor” earned even less at around two shillings per week.Footnote 85 The loans were also significant relative to the price of livestock, such as pigs (£1.50 per head) and sheep (£2.18 per head).Footnote 86
Loan distribution by activity and average loan sizes, 1840
Source: Based on Porter, “Statistical Account of Loan Funds.”
However, there was considerable regional variation in loan use. According to Porter’s account of the activity of 80 LFSs in 1840, the majority of loans were for investment. An average of 40 percent was for agricultural investment, such as livestock or the purchase of seed; 17 percent was for manufacturing, such as for looms and other manufacturing inputs; 15 percent was for dealing (see Table 4).Footnote 87 A significant share of loans, 28 percent, on average, was for consumption, such as payment of rent, clearing debts, and purchasing food, indicating that there were LFS borrowers who required credit for capital investment and borrowers who required credit for day-to-day survival (i.e., income smoothing). And at different times, the same borrower could be both.
Porter concluded that the group that benefited the most from access to loan funds were the “working classes.”Footnote 88 In his summary of evidence, “small farmers” benefited by not having to rely on the much more expensive credit provided by pawnbrokers or private (i.e., unregulated) moneylenders. The Devon Commission also reported that borrowers from LFSs were predominantly small farmers and laborers.Footnote 89 RLFs had similar borrowers: generally small traders and agriculturalists.Footnote 90 The broad range of users and uses of fund societies suggests the credit provided by LFSs was relatively affordable and critical to efficient rural production.Footnote 91 From a borrower’s perspective in Ireland, a Type 1 or Type 2 fund society (as outlined in Table 3) was more beneficial in terms of lower costs of borrowing, but Type 3 and Type 4, although more expensive, may also have been beneficial in terms of accessing credit. LFSs have traditionally been overlooked by historians of Ireland. One exception is Mokyr, who dismisses LFSs in his discussion of sources of credit and summarizes the following conclusion from the Devon Commission: “As loan funds are usually constituted, the highest praise which can truth be allowed to them is, that they are less ruinous than private usurers.”Footnote 92 However, the evidence presented here suggests that a more nuanced interpretation of LFS activity is required.
The LFSs realized profits were applied to charitable purposes or retained. They were seen by Porter as “the property of the poor, though in the hands of trustees for their benefit.”Footnote 93 LFSs were profitable due to the high turnover of capital, frequency of repayment, and the immediate discounting of interest. For example, the Abbeyleix LFS lending averaged £30,000 per annum, and it used its profits to build twenty cottages to rent to laborers; that rental income was then used to fund other charitable ventures, including a fever hospital.Footnote 94 Initially, profits were to fund charitable works, which were substitutes for provisions of the Poor Law. However, regulations evolved stating that profits must be retained as security against losses or to augment capital, thereby providing more funds for the poor to help themselves.
Distributing Profits
LFSs were nonprofits that operated under legislative constraints on business parameters, including caps on interest payable on assets and chargeable on liabilities. They also operated under a constraint that they could not distribute profits. However, they had some autonomy in how net earnings were used: either retained and added to capital or expended for a particular charitable purpose. Legally, funds were not permitted to distribute net profits to members. One way to circumvent this was to increase expenses on activities of indirect benefit to funders or to tie loans to purchases at the shop or pub where the LFS was (illegally) located, thereby reducing net profits or increasing associated returns. How LFSs distributed profits reveals what factors influenced socioeconomics and institutional logics, and how these were affected when the law governing interest chargeable and payable changed in 1843, altering restrictions on the distribution of profits. The “amount expended for charitable purposes” (a column heading in the end-of-year accounts across all LFB books) was a choice that each LFS made independently. This attribute of the LFS paradigm influenced many promotors to advocate for the creation of LFSs because it enabled them, on the one hand, to provide cheap loans and use the “profits” to provide additional charitable benefits. On the other hand, retention of profits created a buffer that protected depositors, made LFSs sustainable, and enabled societies to provide ongoing charity in the form of continued concessional loans to the poor.
Section 24 of the 1836 Loan Society Act stated that profits shall be supplied to charitable purposes such as “in support of the hospital or infirmary of the county” or “for such other local charitable Purpose as they shall think fit; the Residue, or the Whole of such net Profits if no Part shall be so appropriated, being employed as Part of the Funds of the Society until the Trustees shall otherwise determine.”Footnote 95 Alongside reducing interest on deposits and loans, section 44 of the 1843 Loan Fund Act changed the matrix regarding net profits.Footnote 96 It required that some of the funds be set aside as a reserve for the protection of debentures, to augment society’s capital with retained earnings, or to expend profits on charitable purposes “with the Approbation of the said Loan Fund Board.”Footnote 97 Thus, there was a clear distinction between the operational constraints before and after 1843.
LFSs were quite concerned about the reduction in interest rates on their general sustainability and the general air of “uncertainty to which effect the [1843] Act may have” on LFSs’ activity.Footnote 98 Here, we model the choice to spend profits on charitable purposes in 1842 and 1843—that is, immediately before and after the introduction of the 1843 act. The effect of this change was instantaneous. The amount expended on charity dropped by 25 percent, from £18,967 in 1842 to £14,149 in 1843, and the mean amount expended on charity across all funds in 1842 was £45 as compared to £37 in 1843. Modeling the choice to spend profits on charitable purposes reveals the institutional logics of LFSs before the change in the legislation, and how institutional logics themselves changed after the legislation. We analyzed the choice to spend any profits on charitable purposes rather than the amount spent because many LFSs did not spend anything on charity. Thus, even a small amount expended on charity reveals an underlying institutional logic that we distilled down to a simple binary choice: “to give” or “not to give.” The charitable functions of LFSs in the years 1842 and 1843 are analyzed using the following discrete (binary) choice logistic model:Footnote 99
$$ {Log}_e\;\frac{\pi_i\;\left( Charitable\ expenditure\right)}{1-{\pi}_i\;\left( Charitable\ expenditure\right)}=\beta {X}_i+{\varepsilon}_i $$
where charitable expenditure is defined as a binomial that takes the value 1 if an LFS expends money for charitable purposes, and 0 if otherwise. Effectively, the dependent variable is the probability of giving to charity over not giving to charity. The aim of the exercise is to determine whether LFSs’ characteristics are correlated with the propensity to give to charity; the model thus tests for characteristics that might shift this probability up or down.
The independent variables were collected from LFB reports, Poor Law administration reports, and the 1841 census (Table 5; see Table A2.1 for summary statistics). The data from LFB annual reports include information on sizes of LFSs (capital) and income generation (gross income as a percentage of capital), number of depositors, interest paid on capital, and the name of the corresponding officer (which is coded as 1 if it is a religious minister). The number of RLFs and the number of TSBs are included as explanatory variables. Additional controls are included for population density, Poor Law valuation per acre (a proxy for wealth), ratings before 1842, the location of major cities, and the industrial province of Ulster with its established banking system.Footnote 100 It is important to note that logistic models are sensitive to misspecification and will be inconsistent if explanatory variables are omitted. In this regard, one of the most important steps is to assess which function of the dependent variable yields a linear function of the independent variables (the link).Footnote 101 Link tests and other diagnostic tests of the model are reported with the results.Footnote 102 The proportion of the Area Under the Curve (this value compares the sensitivity and specificity of the model, where values range between 0.5 and 1) is used to test the predictive ability of the underlying logistic model. For values at 0.5, these are poor fits and the model may be random. Values between 0.5 < 0.7 are poor, values between 0.7 < 0.8 are acceptable, values between 0.8 < 0.9 are excellent, and values > 0.9 are outstanding.Footnote 103
Definition of all variables used in regression analysis
Sources: LFS data from Loan Fund Board reports: British Parliamentary Papers (BPP), Fifth Annual report; BPP, Sixth Annual report; BPP, Eight Annual report. RLF locations: BPP, Second Annual report. TSB locations: Pratt, Summary of the Savings Banks. Poor Law data: date of first admission of paupers, valuation per acre, rates 1839–1844, or number admitted to workhouse: BPP, Tenth Annual report of the Poor Law Commissioners (Appendices A, Appendix B [Tables 12–15], and Appendix C).
The hypothesis is that fund structures influenced the charitable expenditures of LFSs. First, older LFSs would have had a greater motivation to expend on charitable purposes (or to accumulate greater reserves to be expended), but this would matter less with the change in operational constraints. Second, the inclusion of religious members on the boards of management of fund societies may have influenced charitable decision making. Last, the practical implementation of the Poor Law would have influenced decision making before the rule change but not afterward.
The results of this choice—of whether LFSs expended profits on charity for the years 1842 and 1843—are seen in Table 6 and Figure 4; Figure A2.1 displays results excluding outlier LFSs. However, given that half of the societies chose not to expend money on charity, the focus, again, is on the choice to expend money on charity and not the amount expended itself, which is partly a reflection of the size and profitability of a fund. Results are presented as marginal effects, which are the effect on the conditional mean of y (the choice to make charitable expenditures) of a change in one of the regressors, at the sample mean of the regressors.Footnote 104 The key finding is that, on the one hand, prior to the legislative change, the significant positive predictors of expenditures on charity were the age of a fund, whether a fund had a religious minister on the board of trustees, and the size of a fund (proxied by capital). On the other hand, the implementation of the Poor Law was a counter-balancing force: after the Poor Law was in operation, there was less impetus to expend on charitable services already provided for by ratepayers. Notably, when the legal parameters changed in 1843, there were noticeable effects on behavior. The influences that had led to charitable giving in the past were less influential (e.g., the role played by religious ministers was sizably reduced, and the age of a fund and poor-house admission were no longer significant). Instead, interest paid on capital (indicator of deposits) was a negative predictor of charitable expenditure. The greater the income generation, after the fall in discount rates, the more the number of employees became a positive predictor. There were also contrasting influences of the substitutable and complementary microfinance institutions: RLFs in proximity were negative predictors of charitable expenditure, whereas TSBs in proximity were positive predictors of charitable expenditure.
Logistic regressions of charitable expenditure
Notes: Robust z-statistics in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1; Additional controls: population density, valuation per acre, population density * valuation per acre, dummy for ratings in 1839, 1840, 1841. City dummy (1 = city), Ulster dummy (1 = Ulster). Specification tests: Link test for specification: 1842 z-statistic for ŷ = 6.76 and ŷ 2 = –0.001; 1843 z-statistic for ŷ = 6.21 and ŷ = –0.42. Goodness of fit tests: Hosmer-Lemeshow test statistic is not statistically significant in either year (1842 = 2.58, 1843 = 6.36).
Source: Author’s compilation.
Comparison of marginal effects at means, 1842 and 1843.
Source: Authors’ compilation.
The timing of this observational study could be influenced by the wider business cycle, which had its peak in 1836 and its trough in 1842.Footnote 105 Budgetary changes could also have affected agricultural incomes, especially because the largest decline occurred in 1841 (-2.8 percent), which was followed by another decline in 1842 (˗0.8 percent).Footnote 106 These factors may have influenced LFSs to choose to have more charitable expenditures than they might otherwise have done.Footnote 107 Although difficult to disentangle, an alternative approach to identify the logics of LFSs is to analyze the choices of funds that were continuously active before and after reform. An issue here is the easy entry and exit of LFSs; therefore, the analysis is restricted to funds that were continuously active over three consecutive years (1841, 1842, and 1843), This exercise yields 228 continuously active funds.
Clear distinctions can be seen across the mean of variables associated with the different decision making of LFSs: to never make charitable expenditures; to always make charitable expenditures; to stop, to start, or to repeatedly stop and start (Table 7). A higher proportion of the LFSs that always choose charitable expenditures (93 percent) and those that had given to charity but stopped (91 percent) were formed before 1841. In terms of the Poor Law influence, a high share of the funds that always included charitable expenditures (97 percent) were in PLUs that were not rated before 1841, and they had a lower Poor Law valuation (£0.72 per acre) than the LFS average (£0.85 per acre). This contrasts with LFSs that started charitable expenditures: 88 percent of these funds were in unions not rated before 1841 and which had a significantly higher valuations, on average (£1.10 per acre). LFSs that always gave to charity were also in unions, but where there were fewer poor-house admissions (average, 0.21 percent of the total population of an area). This was half the average of LFSs that never to give to charity (0.48 percent of the total population of an area) or LFSs that started to give to charity (average 0.40 percent of the total population of an area). This suggests possible mechanisms for how the Poor Law influenced LFSs’ decision making on giving to charity: the societies that always made charitable expenditures were located in poorer districts when the Poor Law was in its infancy. In terms of charitable logics, religious ministers were a higher share of managers in funds that always gave to charity (51 percent) and those that started giving to charity (50 percent), while the proportion of religious ministers was lowest in funds that never gave to charity (34 percent). In terms of commercial variables, LFSs that chose to make charitable expenditures had higher values of capital and interest on capital than the LFSs that never gave to charity. The influence of RLFs and LFSs are also notable: more RLFs were found in PLUs in which LFSs never made charitable expenditures (1.43) as compared with LFSs that always made charitable expenditure (0.36). On average, TSBs were found in equal number in unions in which LFSs never or always gave to charity (0.74 and 0.75, respectively). As these are unordered decisions, and there are more than two choices (i.e., no longer binary), the decision can be analyzed using a multinomial logistic regression (see Table A2.2 and Figure A2.2).
Mean of variables by classification of loan funds
Note: Standard deviation is in brackets.
Source: Authors’ compilation.
Following reform of the LFS system in 1843, there was an increase in market exits (see Figure 3). Of the 300 LFSs on the LFB’s register in 1842, 92 had exited the market by 1845 (27 of which had exited by 1843); and of the 298 funds that were on the LFB register in 1843, 73 had exited the market by 1845. Here, exit means either no longer on the register, marked as having “ceased operations,” or having been “gazetted” for irregular activity. By the eve of the famine, a significant number of LFSs had ceased to operate. Applying the logistic regression framework from Equation 1 (above), it is possible to tentatively explore the determinants of the market exit (Figure 5; see full results in Table A2.3), where inactivity (exit) in 1845 is coded as 1 and activity (continuation) is coded as 0. The results highlight the conflictual structure of the LFSs. Variables positively associated with exit are those associated with a charitable logic (e.g., the total sum of charitable lending, religious ministers on the board of management) and associated institutions with similar logics that could fill the void after exit (i.e., RLFs and TSBs). Variables associated with market logics were negatively associated with exit (change in capital, age of fund, number of depositors, and existence of joint stock banks).Footnote 108 The key changes in business parameters following the 1843 act are also revealing. The interest rates paid on capital are positively associated with exit, implying funds paying higher rates of interest found it difficult to operate sustainably after the law reduced margins. Whereas gross income was a negative predictor of exit, here the reduction in discount rates was compensated by the use of fines and ancillary fees, which enabled funds to cushion the blow of the loss of income. For example, the average rate of fines to capital were similar in both 1842 and 1843 (3.41 percent and 3.53 percent, respectively); there was also a 1 percentage point increase in average ratio of income to capital between 1842 and 1843.
Logistic regressions of 1842 and 1843 that ceased operations in 1845, marginal effects at mean.
Source: Authors’ compilation.
To summarize, when regulation was tightened and interest rates were reduced, firms that had operated on purely market logics or had struggled to fulfill their charitable goals exited the market. What is apparent here is that the philanthropic purpose of the LFSs had not changed, but the mechanism for distributing that philanthropy had. Now, instead of direct provision, the provision was indirect via lending. Thus, the more able LFSs could concentrate on their fundamental philanthropic institutional logic: to help the poor help themselves. The Poor Law provided when that help was not enough.
Conclusion
Ireland experienced a revolutionary boom in the provision of microfinance to the poor in the prefamine period of 1836 to 1845. Already staggering under the weight of other taxes, property owners were incensed by the introduction of the Poor Law in 1838. LFSs were seen as a pragmatic private sector means of lessening this fiscal shock by providing the poor with the credit required to escape chronic poverty and offer relief. We examine whether LFSs achieved their desired end. Although descendants of Loan Fund Societies are not a feature of the modern Irish financial landscape, this was an emerging industry of the time, and as much can be learned from its failure as from its success.Footnote 109
LFSs, registered and regulated by the LFB,Footnote 110 were found in the east of the island. RLFs were the second strand of microfinance, and these were predominantly in the west and south of the island. The major source of capital for RLFs were funds raised by the London Relief Committee in 1822, clearly independent of the introduction of the Poor Law in 1838. LFSs were unlikely to enter markets where RLFs were found because the credit needs of the poor were already being met. LFSs were, nevertheless, established throughout the island. They operated in all Poor Law Unions and provided credit to the credit starved. Their target clientele—the “industrious poor”—were mostly small farmers and hucksters (dealers). Both LFSs and RLFs improved this class of people by helping them to better help themselves, at least in normal times.
LFSs and RLFs are like today’s payday lenders and credit unions, which provide credit to the severely credit constrained. While payday loans are expensive, like the LFS and RLF loans, they are cheaper than the alternatives available. Borrowers generally use this option sparingly, even though loans are easy and fast to procure.Footnote 111 The initial regulation of LFSs under the 1838 Loan Societies Act gave societies flexibility to match their institutional logics. Societies with dual charity and credit provision logics, similar to today’s credit unions, used profits for charitable purposes. Societies working solely on market logics, similar to today’s payday lenders (which charge exorbitant interest rates), did not. When regulations were tightened in 1843 by the Charitable Loan Societies (Ireland) Act, reducing interest rates payable on debentures and chargeable on loans, LFSs that had operated on market logics found the new parameters inconsistent with their logics and exited the market. Of the 300 LFSs registered with the LFB in 1842, only 208 were still active in 1845 on the eve of the Irish Famine. Cobb and colleagues argue that in the face of uncertainty, commercial microfinance lenders did not consider exiting because they had dedicated the time to building expertise, developing networks, and investing in physical offices.Footnote 112 This is contrary to the many Irish LFSs that found exiting to be the only viable option. Their behavior is reflected in the similar actions of modern payday lenders in the United Kingdom, such as Wonga, QuikQuid, and The Money Shop. These contemporary lenders have all exited the market because of the tighter consumer protection regulations and interest rate caps promulgated by the Financial Conduct Authority. These were inconsistent with their business models and undermined their profitability.
Other LFSs adapted their logics to the new regulatory reality by continuing to provide loans while sustaining themselves by plowing their profits back into their businesses. This allowed them to fulfill their primary charitable logic of helping the industrious poor to help themselves. Societies founded in this period were still active one hundred years later. There were twenty-five LFSs active in the Irish Free State in 1945. Examples founded a century earlier include the Limerick Industrial and the Limerick Pery and Jubilee; by 1945, they were run solely on accumulated reserves (97 percent and 100 percent, respectively).Footnote 113 The LFSs’ behavior echoes Battilana and Dorado’s argument that balancing and adapting logics is crucial to sustainability of hybrid organizational forms.Footnote 114 Just as the behavior of defunct payday lenders and some LFSs repeat, so too does the behavior of LFSs similar to credit unions in their lending practices.Footnote 115 Credit unions, which are financial institutions concerned with the social and economic success of their communities, face restrictions on the interest rate they can charge. They provide small loans while working with borrowers individually to improve their financial resilience. The long-term benefit to credit unions is higher net returns on lending and improving their own and their community’s sustainability. The long-term benefit to a credit union customer is economic independence, which was precisely the goal of some LFSs in lending to the industrious poor.
Supplementary Materials
To view supplementary material for this article, please visit http://doi.org/10.1017/eso.2020.67.
Open access
