Printer Friendly

Savings banks, famine and financial contagion: Ireland in the 1840s and 1850s.

Savings banks in the United Kingdom lack the glamour of contemporaneous innovations such as the power loom or the railway engine. What could be duller than the story of institutions that did not even expose themselves to the risk of lending money and that, for the most part, were cautiously and competently run? Yet savings banks are worth studying, if less for their own sake than for what they can tell us about economic conditions and economic behaviour more broadly. This article uses the records of a nineteenth-century Irish savings bank to shed some light on the role of financial institutions, and financial panics associated with them, in a poor, backward economy. The outline of the article is as follows. Part I offers a brief introduction to the history of savings banks in Ireland. Part II describes one well documented bank, the Thurles Savings Bank. In Parts III and IV the impact of two financial panics on the Thurles Savings Bank is discussed. Part V concludes.

I

During the industrial revolution there was no shortage of schemes directed not only at 'industrious and frugal' servants and tradesmen, but more generally at those who might easily be reduced to destitution by unemployment, illness or old age. In 1793 the British parliament passed a scheme to promote friendly societies, but such societies were soon being criticised for being wasteful and too narrowly focused. Of several schemes aimed at encouraging working-class thrift the most important would prove to be the provident institution or trustee savings bank. The foundation of the savings bank movement is usually dated to 1810 in Ruthwell, a small village in lowland Scotland. The new concept spread very rapidly throughout the United Kingdom and farther afield. (1) It became

Special thanks to James O'Shea for the database of Thurles Savings Bank account holders. The comments of two anonymous readers are also gratefully acknowledged. fashionable for the middle classes and the gentry to be involved in banks as trustees, patrons or part-time managers. Their desire to make the poor industrious was coupled with a self-interested concern to reduce the nuisances of poor relief and street begging. The system thus embodied a paternalism that seemed to unite the interest of rich and poor.

The new institutions, relying on a combination of public and private subsidy, aimed to offer their humble clients three things: a relatively attractive return on their savings, considerable liquidity and security. So influential was the support for them in the United Kingdom that parliamentary backing was soon forthcoming. Separate Acts to encourage the spread of savings banks in Ireland and in England fixed the rate of interest payable to depositors at a very generous 4.55 per cent per annum, limited depositors to investments of 50 [pounds sterling] per annum in Ireland and 100 [pounds sterling] in Britain, and exempted bank transactions from stamp duty. (2) They also prohibited trustees from having a financial interest in a savings bank. As a confidence-building measure the legislation further stipulated that the banks' deposits be placed on account with the Commissioners for the Reduction of the National Debt.

Some radicals, for example William Cobbett, viewed savings banks as a cunning device for getting rid of the entitlement to poor relief. While undoubtedly some supporters of savings banks were hard-line Malthusians who wanted to put an end to all poor relief, most would have recognised the validity of entitlements to some kind of relief. More soundly based was the critique that the banks were really not helping those for whom they were intended, and that the main beneficiaries were the middle and lower middle classes. This point would be rediscovered in the 1950s and 1960s by sociologist Neil Smelser and economic historian Albert Fishlow. (3)

In the mid-1810s the new banks spread like wildfire. By the end of 1818 there were nearly 500 savings banks in Great Britain. Diffusion tapered off thereafter. In Ireland diffusion lagged, but only very slightly, behind the rest of the UK. (4) As in Britain the banks relied on local grandees to lend prestige, and on clergymen, professionals and businessmen to provide the initiative and to act as trustees or managers. This sense of noblesse oblige (or its bourgeois equivalent) was a crucial aspect. In general the management was ecumenical in composition.

Ireland's first successful savings bank, the Belfast Savings Bank, opened for business in January 1816. (5) Thereafter, although Ulster took the lead, banks were soon set up throughout the island. By late 1829 there were seventy-three of them, and on the eve of the Famine there were nearly 100,000 depositors in seventy-six savings banks holding balances totalling nearly 3 [pounds sterling] million. The Irish savings bank network had been essentially established by the mid-1820s. Of the seventy-four Irish banks still open in late 1846, forty-six had been created in 1816-25, a further twenty-one in 1826-35, and only seven from 1836 on. Long-established banks best withstood the pressures of the late 1840s (on which more later). Of the fifty-two founded before 1826 six had disappeared by 1848; of the twenty-nine founded in 1826-35, eight had failed by 1848; of the twelve founded in 1836 or later, five had folded by 1848. The earlier savings banks were also bigger.

On the eve of the Famine the population of Ireland was more than half that of England and Wales, and more than double that of Scotland. Yet Ireland had only half as many savings banks as Scotland, and about one-sixth as many as England and Wales. Part of the reason for this was that banks fared best in urban settings, whereas Ireland was overwhelmingly rural. The banks were more likely to be located in the more developed parts of the country. On the eve of the Famine the province of Connacht, poorest and least urbanised and worst affected by the Famine, had 17 per cent of the population but only 4 per cent of the savings held in savings banks. So, ironically, the banks were fewest where the really poor were most numerous. The initial impetus behind the banks, and the main concern of their parliamentary supporters, was to get the poor to save. In this respect their record was mixed at best. Hard data on the savers' socio-economic status shows that the lower middle and middle classes were the main beneficiaries. (6)

II

Thurles Savings Bank (henceforth ThSB) opened for business in late 1829 and folded in 1871. (7) The decision to create the bank was taken at a meeting of 'those gentlemen who are disposed to lend their aid ... for the benefit of the town and neighbourhood'. Its trustees and managers were mainly local clergymen, landed proprietors and professional people, most of whom were long-serving. Between 1829 and 1859 the bank had only three treasurers (after which the National Bank fulfilled the function), and local shopkeeper, stamp seller and stationer Matthew Quinlan served as part-time actuary from beginning to end, on a salary that varied with the volume of business. However, only a minority of the twenty trustees nominated at the outset played any significant part in the ThSB's operations, and some seem never to have attended a quarterly trustees' meeting. 8 The bank was run by a small, conscientious group of local men who did not, however, overly concern themselves with the original mission of savings banks, i.e. helping the very poor to save.

Uniquely for Ireland, the records of the ThSB survive almost in their entirety. (9) The bank was representative of banks in middle-size towns like Thurles, though the average sum deposited was slightly on the high side, perhaps because the ThSB was more willing to accept multiple deposits from single households. A feature was the particularly high percentage of savers in the 20 [pounds sterling]-50 [pounds sterling] bracket: 52 per cent of all accounts in Thurles, against 38 per cent nationally in November 1846. (10)

Throughout its lifetime the bank opened only once a week, on Mondays between 1.00 p.m. and 2.00 p.m. Between 1829 and 1871 it received 187,057 [pounds sterling] 10s 6d in 4,213 accounts opened in the names of individuals, as well as another fifty-one representing voluntary organisations or charitable institutions. (11) Male accounts outnumbered female, though not strikingly so (by 2,387 to 1,826). The average opening balance in male accounts exceeded that in female by 19 [pounds sterling] 14s to 17 8s [pounds sterling]: a very slender margin, given the big gender gap in earnings in nineteenth-century Ireland. Since the number of transactions per account was small, a significant share of the withdrawing and depositing of money was done through opening and closing accounts.

The ThSB's ledgers corroborate the suggestion that the banks were mainly a vehicle for the more comfortable and better-off. The spread of opening lodgements is worth remarking on (Figure 1). More than one lodgement in three (1,630 out of 4,213) was for exactly the maximum permitted sum of 30 [pounds sterling]. Note too the peaks at 5 [pounds sterling], 10 [pounds sterling] and 20 [pounds sterling]. Quite plainly children's accounts were used to overcome the regulation that no single account be augmented by more than 30 [pounds sterling] in a single year, and the bank's managers indulged this practice. The opening deposits in trust accounts, i.e. accounts held in the names of others, mainly minors, tended to be bigger than average. Only 8.5 per cent of them were of 5 [pounds sterling] or under, compared with 18.5 per cent of all opening deposits. Moreover, nearly three-fifths of the opening deposits of exactly 30 [pounds sterling] were trust accounts, and a much higher proportion of trust accounts (52.6 per cent) were at the upper limit of 30 [pounds sterling].

Between 1829 and the beginnings of the Great Famine in 1846, deposits in the ThSB exceeded withdrawals in each year, with the exception of 1840 and 1842. However, in 1834-36 there were substantial withdrawals (11,265 [pounds sterling] against 14,340 [pounds sterling] deposited). This may well have been due to the opening of a branch of the National Bank in the town in 1835, and of a branch of the Agricultural & Commercial Bank in the following year. The opening of a branch of the new Tipperary Joint Stock Bank (on which more later) in 1840 also drew some accounts away from the ThSB. As in England, reductions in the rate of interest paid on deposits in savings banks also probably contributed something. The sensitivity of deposits to alternative outlets for funds is significant.

[FIGURE 1 OMITTED]

An earlier analysis of Thurles savers (12) produced some expected and some perhaps surprising features. It showed that unskilled labourers (males, typically employed on a daily or weekly basis in pre-Famine Ireland) and servants (mostly female in the ThSB's catchment area, and likely to be employed for months at a time) were underrepresented. Labourers accounted for half the labour force on the eve of the Great Famine, but for only one saver in fifty. In effect the ThSB was a bank for farmers and their families. More than one account holder in four was described as a farmer or a member of a farming family, and it is clear from the ledgers that a significant number of those described merely as 'minors', 'spinsters', 'widows' and 'married women' were also from farming families. These categories were to the fore throughout the bank's history. Savers opening their accounts with a deposit of less than 2 [pounds sterling] included three labourers, thirtyeight servants, seven bakers and two farmers. Savers opening with an even 30 [pounds sterling] included seven labourers, eight servants, one baker, 311 farmers and 296 other members identified as belonging to farming households.

The low average opening balances of domestic servants and unskilled labourers are expected, those of tailors and bakers perhaps less so. At the other end of the spectrum were landowners and gentlemen, the groups with the highest average maximum balance. The closeness of opening, closing and maximum balances for farmers, farmers' wives, and farmers' children suggests that farmers used the accounts of family members to extract maximum benefit from the bank.

In general the picture is of rather inactive accounts, with an average of just a few transactions a year, and the number of deposits exceeding withdrawals. This seems to have been typical of nineteenth-century savings banks. (13) In the pre-Famine period, when the ThSB was most active, the average closing balance exceeded the average opening balance in all occupational categories. This suggests that the bank was used as a vehicle for accumulation. The average account was held for about four years, with little variation here across occupations or parishes. However, it was not unknown for account holders to close their accounts and reopen another later. The significant proportion of accounts in the names of children (11 per cent) and juveniles (12 per cent) again suggests that these were used to circumvent the rules.

Table 1 presents a simple econometric analysis of saving patterns in preFamine Thurles. The dependent variables are OPBAL (opening balance), SAVED (net sum saved), MAX (maximum deposit), DEPS (number of deposits), and WITHDRAWALS (number of withdrawals). These are regressed on the following explanatory variables: thurl means an address in the town of Thurles; loesq includes landowners and those described as 'esq.'; ric represents members of the Royal Irish Constabulary; servant includes domestic servants; lab labourers; and marrfem married females. The variables spinster and minor are selfexplanatory; trust refers to accounts held in trust, while mintrust, spintrst and farfamtr refer to accounts held in trust on behalf of minors, spinsters or farm family members. The coefficients in the columns represent marginal effects. Thus being female, a spinster or a minor was associated with bigger than average opening balances (by 2 [pounds sterling] 7s, 3 [pounds sterling] 8s and 6 [pounds sterling] 2s, respectively), while being a clergyman or a member of a farming family (other than a farmer) meant a smaller opening balance. Trust accounts were bigger and the maximum amount held in them tended to be bigger. Trust accounts held in the names of minors (mintrst), farming family members (farfamtr) and spinsters (spintrst) were used to save (as reflected in the positive coefficients in the SAVED column).

III

The ThSB was subject to two serious runs, in 1848 and in 1856. It is very tempting to link the first of these with the Great Famine. Yet although the Famine undoubtedly affected Irish savings banks, the link between it and the banks' fortunes in the late 1840s is not straightforward. Our profile of account holders would imply that they were less likely to be at risk than the rural, landless poor. Indeed, in the early stages some observers suggested that the savings banks' seeming prosperity belied Irish claims of hardship and crisis. Editorials highlighted reports from Ireland of increases in deposits as evidence of 'successful swindling' or welfare fraud on the part of the people. (14) However, both aggregate data and individual case studies would seem to confirm that the economic shock caused by the Famine dealt a serious blow to savings banks. Between 1845 and 1849 aggregate deposits fell by more than half, from nearly 2.9 [pounds sterling] million to 1.2 [pounds sterling] million, and the number of depositors also dropped by more than half. Of the forty-four savings banks in the United Kingdom that ceased business between 1844 and 1852, twenty-four were Irish. (15)

The Famine placed all Irish financial institutions under pressure, and the savings banks were not immune. But financial contagion, rather than the Famine per se, was responsible for a major run against the country's savings banks in mid-1848. (16) That run led in turn to a huge decline in deposits and account numbers from which the banks never truly recovered. It exposed a weakness in their original design. In Ireland the record of the savings banks in this respect had hitherto been quite good: only one serious case of embezzlement came to light before the late 1840s. But the unrelated failures of three savings banks in 1848 were enough to give rise to a systemic panic. The savings banks would not have remained immune from the economic crisis associated with the Famine, but it was the unrelated, panic-induced contagion of 1848 that almost destroyed them.

Oliver Horne, historian of savings banks, described the failure of St Peter's Parish Savings Bank, located on Dublin's Cuffe Street, as 'the first real sign of a chink in the armour designed by parliament'. (17) That bank, established in 1818 with the backing of several eminent trustees, had been the target of embezzlement and mismanagement by its cashier since the mid-1820s. A run on the bank in November 1845 marked the beginning of the end. When the Cuffe Street bank finally closed its doors on 10 May 1848 its liabilities had reached nearly 65,000 [pounds sterling] against assets of 100 [pounds sterling] or so. Sensing that the game was up and that compensation was unlikely, some depositors began to sell their passbooks at a discount the following week. (18)

Much more damaging were the sensational, unrelated collapses of the Tralee and Killarney savings banks in April 1848. (19) The Tralee bank operated from its actuary's own dwelling place, 'which afforded him considerable latitude for carrying on his frauds'. Since depositors called at all hours with their deposits there were no managers present to check entries. One of the actuary's scams worked as follows. Deposits of 30 [pounds sterling], 15 [pounds sterling] and 27 [pounds sterling] would be entered as 3 [pounds sterling], 5 [pounds sterling] and 7 [pounds sterling], and a sum of 15 [pounds sterling] added to the coffers. The manager would see that the sum lodged matched the entries. Then the actuary would add a zero to the 3 [pounds sterling] and change the 5 [pounds sterling] to 15 [pounds sterling] and the 7 [pounds sterling] to 27 [pounds sterling], so that depositors who came to claim their money would get it. In this way suspicions were not aroused. The actuary, on a salary of 60 [pounds sterling], pocketed about 28,000 [pounds sterling] between 1832 and 1848, although in 1848 'he appeared to have had but 3,000 [pounds sterling] realised'. He had built up liabilities of 36,768 [pounds sterling] against 1,650 [pounds sterling] assets, for which he got fourteen years' transportation. (20) The Killarney Savings Bank, where over 1,000 savers held accounts, closed its doors on 18 April 1848. In this case the actuary managed to escape, leaving liabilities of 36,000 [pounds sterling] against assets of 16,582 [pounds sterling].

An official arbitrator, John Tidd Pratt, was appointed to look into the plight of the two Kerry banks in May 1848. (21) His findings were highly critical of the banks' management, but he found 'the greatest abuse [had] existed on the part of the depositors, with respect to their mode of depositing, and the amounts they invested'. His revelations created a sensation. Irish politicians and the press sympathised with the depositors and berated Tidd Pratt for protecting the rich (i.e. the trustees) at the expense of the poor (the account holders). Tidd Pratt countered in a letter from Killarney to the Freeman's Journal in which he revealed that the average deposit in Tralee had been 40 [pounds sterling] 'and in this place will exceed 50' [pounds sterling]. He revealed that as numerous Tralee account holders handed in their pass-books to the clerk in the hope of getting some of their money back it had emerged that some apparently impecunious farmers had huge sums lodged, 'even over a thousand pounds each', while others had claimed sums of hundreds of pounds. (22) In no savings bank in the United Kingdom had he ever found 'so great a number of what I consider large accounts'. It was no different in the case of the Killarney savings bank: 'tenants, who pleaded extreme poverty to their landlords, paupers from the workhouse, and men whose outward appearance would lead you to look on them as objects of charity, were soon at the office door'. (23) Tidd Pratt reduced the claims of all depositors to those allowed by the rules, thereby drastically cutting the bill facing the trustees, from perhaps 40,000 [pounds sterling] to about 10,000 [pounds sterling]. He added that his duty was 'far from being a pleasant one'. (24)

In amusing evidence to a House of Commons select committee on savings banks in 1849 Tidd Pratt described cases 'where husbands brought books representing the money to be the property of their sisters, and upon calling the sisters it turned out to be their wives', and of people producing account books which they claimed were not their own but those of their unsuspecting nephews and nieces. (25) Tidd Pratt's irritation at what he deemed 'the utter disregard of truth, the falsehood and subornation of perjury' displayed by the Kerry claimants was understandable. Yet he was perhaps too ready to accept the assertions of some of his friendlier informants as fact. (26)

Tidd Pratt's sensational accusations were made public on 18 May 1848. Widely circulated in the domestic and foreign press, (27) they were vigorously rebutted by Henry Arthur Herbert, owner of the Muckross estate and recently elected MP for Kerry. Herbert produced evidence from the prison governor to refute the damaging claim that three men who had been jailed for debt 'had presented themselves in custody of their gaolers to claim as depositors'. Tidd Pratt was forced to withdraw that particular accusation before a committee of the House of Commons in 1849. (28) Another widely circulated claim, that inmates had left Killarney workhouse in search of their deposits, was also probably a fiction. Herbert was given the names of four workhouse inmates who, according to the master, had applied for dismissal at the time of Tidd Pratt's hearings, and whom fellow inmates had jokingly accused of being depositors, but a search yielded no matching names. (29)

The failures in Dublin and Kerry resulted in country-wide financial contagion. In Cork the trustees of the local bank were forced to withdraw 45,000 [pounds sterling] of their investment in the national debt during the first half of April 1848 in order to meet a serious run. Not even Belfast, less affected by the Famine than most places in Ireland, was immune. (30) In Thurles the disaster resulted in more accounts being closed in 1848 than in any other year in the ThSB's history. Depositors were slow to return to the savings banks, and recovery was impeded by a more aggressive search for accounts on the part of the joint-stock banks after mid-century. Cashing in on the panic, the National Bank began to accept deposits of 10s or more at the current rate of interest. The fragility of the savings banks after 1848 is well reflected in the impact of a rumour a few years later that the Cork Savings Bank had closed for good, when in fact it was merely refurbishing its facilities: the result was a serious run on that and other banks. (31)

Were those who panicked in 1848 systematically different from those who held their nerve? (32) The archive of the ThSB sheds some light on this question. In an earlier paper (33) I compared the panickers (approximated by those who closed accounts between April and September 1848) with four other sets of Thurles account holders: first, those who closed their accounts in 1843-45; second, those who closed between January 1847 and March 1848; third, those who closed in 1849-51; and finally, those who held accounts in March 1848 but chose not to close them in the following months. The outcome implied the absence of any strong gender affect, and the opening and closing balances of those who panicked differed little from the balances of those who did not. Account holders with addresses in the town of Thurles were slightly more inclined to panic than those living outside the town, but again the effect was small, and there was little evidence, either, of panickers clustering by parish. Two differences were more significant. During the panic account holders with the same surname and address were more likely to close.

Farmers and members of farming households were also more likely to close, while people of means, such as landowners, clergy and professionals, were less likely to do so. Note too that servants and labourers were also more likely to keep their accounts open, though only marginally so. It is hardly surprising that, when parents closed accounts, they were also likely to close those of their children. That networks of occupation or parish did not register is perhaps more surprising, to the extent that account holders from a given area or employment might have been expected to compare notes.

Table 2 offers a more formal analysis. It describes the outcome of four slightly different logit specifications of those who 'panicked'. The dependent variable is set at unity for accounts closed during the panic period, and zero otherwise. The new explanatory variables are as follows: inner and outer refer to addresses in an inner and outer ring of parishes around Thurles; relpanic is set at 1 if the account of a relation is closed during the panic; whcollar includes landowners, those described as 'esq.', corn dealers and professionals; ric represents members of the Royal Irish Constabulary; prole includes mainly labourers and servants. The other covariates refer to particulars of each account: dateop (when the account opened); avl (lodgements per day); avw (withdrawals per day); 14548 (lodgements in 1845-48); w4548 (withdrawals in 1845-48); opbal (opening balance); op45 (balance in 1845); c148 (balance in 1848); totdep (total sum deposited); and max (maximum deposit). The coefficients in Table 2 are the marginal effects of a change in each variable, and the associated t statistics. The unimportance of gender is confirmed. Nor did location matter much; indeed, those living in the outer ring of parishes were most likely to panic. Farmers and unskilled workers were more panic-prone, members of the police force and the wealthy less so. The longer an account had been open the less likely it was to close: perhaps some accounts were effectively dormant in 1848. Inertia seems to have been a factor: the busier an account, as reflected in avl and avw, the more likely it was to close. The more on deposit in the account in 1848 (c148) the less likely it was to close. Big accounts, represented by max and totdep, were also less likely to close. This is consistent with the hypothesis that the less well-off among the depositors were more inclined to panic.

The strong tendency for family members to act together is striking. This is captured by the big and significant coefficient on relpanic throughout. One possible interpretation of this is the importance of the family as an informational network. In other words, in Thurles the decisions of siblings or other family members mattered more than those of neighbours. But there is also a more mundane interpretation for this pattern: in many cases trust accounts in the names of sons and daughters were managed by parents, and indeed used as a means of circumventing the rules about maximum lodgements and sums deposited. This means that relpanic's status as a truly independent variable is suspect.

The panic of 1848 had a lasting effect on Irish savings banks. As the joint-stock banks expanded, the savings banks foundered. Deposits would never recover their pre-panic level. The failure of the savings banks exposed a shortcoming in their design, and undermined popular confidence in them. In all cases the trustees had failed in their responsibility to monitor the bank's actuary. The abuse of the system highlighted by Tidd Pratt made those with substantial deposits in the savings banks doubly nervous. There is thus a sense in which both the panic and the subsequent reluctance to use savings banks were 'rational' reactions. However, it is also noteworthy that an equally sensational collapse less than two years later in England did not have the same effect. On the death of George Haworth, actuary of the Rochdale Savings Bank, in November 1849, it was discovered that he had defrauded depositors of over 70,000 [pounds sterling]. Haworth had been using the bank to fund his foundering cotton mill. Despite this and a series of other smaller defalcations the aggregate sum deposited in English savings banks continued to rise during the 1850s, as aggregate deposits in Ireland fell. (34)

IV

Tipperary in 1856 may have seemed a long way from the London of Charles Dickens and Anthony Trollope. Yet Mr Merdle in Little Dorrit (1857) and, to a lesser extent, Mr Melmotte in The Way We Live Now (1875) were modelled on a notorious real-life Tipperaryman. Like Merdle and Melmotte, John Sadleir, MP, committed suicide in the face of financial scandals. In the last decade of his life Sadleir, a controversial figure in Irish politics, lived mostly in London, whence he exercised full control over the Tipperary Joint Stock Bank, established in 1840 by his uncle James Scully, and managed by his brother James. John Sadleir's suicide on Hampstead Heath on the night of 16-17 February 1856 was prompted by the refusal of both Messrs Glynn in London and the Bank of Ireland in Dublin to cash any further drafts on the Tipperary Bank, and by the imminent disclosure of Sadleir's role in various embezzlement schemes. In the following days the inquest into Sadleir's death and revelations about his various scams figured prominently in the British and Irish press. (35)

The sensational failure of the Tipperary Bank, with liabilities of 0.4 [pounds sterling] million against assets of about 45,000 [pounds sterling], had far-reaching financial ramifications in Ireland. Most Irish banks came under pressure in the following months, though by the end of the year the crisis was over. (36) In Thurles, where the business of the local branch of the Tipperary Bank had been 'rather extensive', both the National Bank and ThSB came under some pressure in the weeks following the collapse. In the immediate aftermath of the collapse a police presence was required to protect the Tipperary Bank's premises against angry farmers. A local newspaper, however, claimed that the panic did not extend beyond 'the small farmer class'. (37)

Were there any discernible differences between those people driven to close their accounts in the ThSB in February and March 1856 and those who closed in 'normal' times? Table 3 compares the seventy-five accounts closed during February and March 1856 with the 199 closed in 1853-55 and the 191 closed in 1857-58. The profiles are quite similar in most respects. However, average opening and closing balances were both higher during the panic than before it; farmers, members of farming families and labourers were much more prominent among closers in 1856 than either before or after; and those who withdrew during the panic were much more likely to be people with the same surname as other closers. Policemen, landowners, professional people and the gentry were less inclined to panic. This is consistent with the suggestion that family networks were, as before, an influence on the decision to close an account.

The collapse was unfortunate for the ThSB in another respect. For many years it had held a balance of several hundred pounds with the National Bank, money that should technically have been deposited with the National Debt Office. When the National Bank announced in mid-1855 that it was about to reduce the rate of interest on those deposits from 2.5 per cent to 2.0 per cent, whereas the Tipperary Bank guaranteed 2.5 per cent, the account was moved to Sadleir's bank. (38) The decision, which cost the ThSB nearly 5,000 [pounds sterling], would haunt it till the end. It forced the trustees in November 1858 to reduce the actuary's salary by 10 [pounds sterling] and in May 1859 to reduce the rate of interest payable on deposits to 2.5 per cent. (39) Had it been widely known that shortly before the collapse the trustees of the ThSB had transferred their account from the National Bank to the Tipperary Bank, the run on the ThSB would surely have been more sustained.

V

The main findings of this paper are twofold. First, the early history of savings banks in Ireland is the story of an institutional innovation that, regardless of the intentions of its originators and supporters, and to a greater extent than in England and even more so than in Scotland, benefited disproportionately the relatively better-off. While many people of modest means undoubtedly benefited from the banks, it is clear that the lion's share of the gains went to a minority of relatively comfortable account holders, attracted by the generous interest rates on offer. For such savers the savings banks represented a lucrative and seemingly secure alternative to commercial banks. The paradox is that in Ireland, where the unskilled and unlettered poor needed an institution such as the savings banks most, they benefited least from them.

Secondly, had other events not intervened, the socio-economic profile of account holders would probably have shielded the savings banks from the worst effects of the Great Famine. (40) Households headed by farmers and policemen, artisans and tradesmen--representative savings bank account holders--suffered too, but were much less at risk of bankruptcy or destitution than those of, say, landless and semi-landless labourers or fishermen. So, surprisingly perhaps, the worst crisis ever to affect Irish savings banks was provoked less by the distress of account holders than by the independent but near simultaneous failure of three poorly run savings banks in 1848. Nonetheless, the Irish economy's relative backwardness may partly explain the savings banks' vulnerability to poor management and embezzlement, and why these failures would prove so damaging to the entire Irish savings bank system.

Cormac O Grada

University College, Dublin

(1) The classic source on the history of savings banks in Britain is H. O. Horne, A History of Savings Banks (Oxford, 1947); but see too M. Moss and I. Russell, An Invaluable Treasure: The History of the Trustee Savings Banks (London, 1994). On Germany and Spain see T. W. Guinnane, 'Delegated Monitors, Large and Small: Germany's Banking System, 1800-1914', Journal of Economic Literature, 40 (2001), 73-124; A. P. Martinez Soto, 'Las cajas de ahorros espanolas en el siglo XIX: entre la beneficencia y la integracion en el sistema financiero', Rivista de historia economica, 18, 3 (2000), 585-628.

(2) These rates and limits would be reduced later.

(3) N. Smelser, Social Change in the Industrial Revolution (London, 1959); A. Fishlow, 'The Trustee Savings Banks, 1817-1861', Journal of Economic History, 21 (1961), 26-40.

(4) C. O Grada, 'Savings Banks as an Institutional Import: the Case of Nineteenth-Century Ireland', Financial History Review, 10 (2003), 31-55; id., 'The Early History of Irish Savings Banks', UCD Centre for Economic Research Working Paper 2008/3 (available at: http://www.ucd.ie/ economics/research/research_papers_2008.htm).

(5) W. E. Tyrrell, History of the Belfast Savings Bank (Belfast, 1946); A. McCreary, By All Accounts: a History of Trustee Savings Banks in Northern Ireland (Antrim, 1991).

(6) O Grada, 'Early History'.

(7) J. O'Shea, 'Thurles Savings Bank, 1829-1971', in W. Corbett and W. Nolan (eds), Thurles, the Cathedral Town: Essays in honour of Archbishop Thomas Morris (Dublin, 1989), pp. 93-117.

(8) Ibid., pp. 96-7; ThSB Board Minutes (Thurles Public Library).

(9) They are held in Thurles Public Library.

(10) Thom's Irish Almanac and Official Directory for the Year 1848, p. 194; ThSB records.

(11) O Grada, 'An Institutional Import'.

(13) Compare P. L. Payne, 'The Savings Bank of Glasgow, 1836-1914', in P. L. Payne (ed.), Studies in Scottish Business History (London, 1966), pp. 152-86; P. L. Payne and L. E. Davis, The Savings Bank of Baltimore, 1818-1866: a Historical and Analytical Study (Baltimore, MD, 1956); G. Alter, C. Goldin and E. Rotella, 'The Savings of Ordinary Americans: the Philadelphia Saving Fund Society in the mid-Nineteenth Century', Journal of Economic History, 54, 4 (1994), 735-67; C. O Grada, 'The Great Famine, the New York Irish, and their Bank', in id., Ireland's Great Famine: Interdisciplinary Perspectives (Dublin, 2006), pp. 184-90.

(14) Rev. J. O'Rourke, The History of the Great irish Famine of 1847 (Dublin, 1902), pp. 214-15; G. L. Smyth, ireland, Historical and Statistical (London, 1844-49), III, p. 29.

(15) James O'Shea, 'Thurles Savings Bank'; Cormac O Grada, Black '47 and Beyond: the Great irish Famine in History, Economy, and Memory (Princeton, NJ, 1999), p. 156; Thom's irish Almanac, 1850, p. 195; ibid., 1851, p. 264; Return of Savings Banks in the United Kingdom that have Failed, Stopped Payment or been Discontinued, Parliamentary Papers 1852 (471) 28.

(16) The following six paragraphs are based in part on O Grada, 'An Institutional Import', 49-54.

(17) Horne, History of Savings Banks, p. 121.

(18) See Report from the (Second) Select Committee on Savings Banks, Parliamentary Papers 1850 (649) 19, appendix, p. 110; Anon. [W. H. Porter], 'Savings Banks: a Review of Papers dealing with Savings Banks', Dublin University Magazine, 34 (1849), 133; W. Neilson Hancock, What are the Duties of the Public with Respect to Charitable Savings Banks? (Dublin, 1852), pp. 6-12; Horne, History of Savings Banks, pp. 119-21; O Grada, 'Savings Banks as an Institutional Import', 50.

(19) Tralee Chronicle, 8 April 1848.

(20) [Porter], 'Savings Banks', p. 131.

(21) Tidd Pratt was long-serving consultant barrister to the National Debt Commissioners.

(22) Kerry Examiner, 11 April 1848.

(23) Ibid., 28 April 1848.

(24) Freeman's Journal, 3 May 1848; Tidd Pratt, cited in Tralee Chronicle, 13 June 1848.

(25) Report of [First] Select Committee on Savings Banks, Parliamentary Papers, 1849 (437) 14, evidence of J. T. Pratt (qq. 3760, 3765, 3776).

(26) Hansard, 3rd ser., vol. 104 (29 March 1849), cols 23-4.

(27) And repeated later--see, e.g., M. Dillon, The History and Development of Banking in Ireland (London, 1889), pp. 104-5.

(28) Report of [First] Select Committee on Savings Banks, q. 3776 (J. T. Pratt).

(29) Hansard, 3rd ser., vol. 104 (29 March 1849), cols 23, 34-5; Report of [First] Select Committee on Savings Banks, q. 3541 (H. A. Herbert).

(30) Tyrrell, History of the Belfast Savings Bank, p. 45; Freeman's Journal, 15 April 1848 (citing the Southern Reporter).

(31) Dillon, Banking in Ireland, pp. 105, 107.

(32) The same question is addressed in the very different context of New York City in the 1850s in C. O Grada and E. N. White, 'The Panics of 1854 and 1857: a View from the Emigrant Industrial Savings Bank', Journal of Economic History, 63, 1 (2003), 213-40; M. Kelly and C. O Grada, 'Market Contagion: Evidence from the Panics of 1854 and 1857', American Economic Re-view, 90, 5 (2000), 1110-24.

(33) O Grada, 'Savings Banks as an Institutional Import', 52-4.

(34) Horne, History of Savings Banks, pp. 126-8.

(35) J. O'Shea, Prince of Swindlers: John Sadleir, M.P., 1813-1856 (Dublin, 1999), especially pp. 379-414, 498-99.

(36) O'Shea, Prince of Swindlers; J. O'Brien, 'Sadleir's Bank', Journal of the Cork Historical and Archaeological Society, 82 (1977), 33-8; F. H. Hall, The Bank of Ireland, 1783-1946 (Dublin, 1947), pp. 229-31, 246-9; Freeman's Journal, 19, 22 February, 5 March 1856.

(37) Tipperary Leader, 23 February 1856.

(38) ThSB Trustees' Minutes, 4 August 1855.

(39) Ibid., 21 November 1858, 18 January 1859.

(40) O Grada, Black '47 and Beyond, chapter 4.
Table 1 Saving Patterns in the Thurles Savings
Bank, 1829-1846

Dependent            OPBAL       SAVED         MAX          DEPOS-
variable                                                    ITS

Adjusted [R.sup.2]   0.111       0.017         0.043        0.061
No.                  2269        2269          2265         2264

Variable             dy/dx       dy/dx         dy/dx        dy/dx

thurles              -1.603 **    2.106 *      -2.746 *      0.3409
female                2.729 **    0.5196        3.068 *     -0.217
farmer               -0.3781     -2.563 **     -4.48 **      1.388 **
farmfam              -3.239 **    0.2552       -3.915        5.112 **
widow                -1.938      -0.2191       -4.027        0.2053
marrfem               1.628       0.2956        0.6496      -0.580
minor                 6.176 **   -2.697         9.027 **    -0.603
spinstr               3.819 **   -0.1197        2.948       -0.459
lab                  -1.255      -0.1898       -4.364       -0.150
clergy               -7.466 **   -6.229        -15.68 *      3.591 *
ric                   1.104      -5.058        -6.116       -0.568
servant               1.78 *      1.191         5.863 *     -0.135
loesq                 0.9483     -3.351        -0.7485      -0.842
dateopen              0.0003     -0.0013 **    -0.0020 **   -0.000
trust                 5.296 **    1.146         4.552 **    -1.78 **
mintrst              -4.577 **    12.236 **     1.863        0.4524
spintrst             -3.642 **    3.1333       -1.103        0.8669
farfamtr              2.486 *     4.1767        9.998 **    -3.23 **

Dependent            WiTH-       Mean
variable             DRAWALS     Value of
                                 Variable

Adjusted [R.sup.2]   0.046
No.                  2269

Variable             dy/dx

thurles              -0.091      0.3784
female               -0.078      0.4159
farmer                0.1760     0.1744
farmfam               0.1979     0.1223
widow                 0.9287 *   0.0336
marrfem              -0.560      0.0653
minor                -0.759      0.0861
spinstr              -0.891 *    0.0706
lab                  -1.16 **    0.0234
clergy                1.892 *    0.0071
ric                  -0.566      0.0247
servant               0.5649     0.0583
loesq                -0.406      0.0238
dateopen             -0.000      9.1838
trust                -1.45 **    0.4309
mintrst               0.0992     0.0428
spintrst              0.7810     0.0309
farfamtr             -0.092      0.0490

Note ** Statistically significant at 5%. *
Statistically significant at 1%.

Table 2 A logit model of who panicked in 1848: marginal effects

Number of observations         836                 838

Probability > [chi.sup.2]      0.000               0.000
Log likelihood              -444.776            -445.2124
Pseudo [R.sup.2]               0.220               0.221

Variable                    dy/dx       Z       dy/dx       Z

sex                          0.0057      0.14
thud*                        0.0427      0.40
inner*                       0.1442      1.35     .1066      2.19
outer*                       0.0838      0.72     .0450      0.71
relpanic*                    0.4668      9.15     .4685      9.19
whcollar*                   -0.0440     -0.42   -0.0446     -0.44
ric*
farming*                     0.0823      1.57    0.0782      1.55
prole*                       0.0668      0.91    0.0673      0.92
spinster*
minor*
dateop                       -.0000     -1.64    -.0000     -1.80
avl                         96.40        6.15   96.82        6.30
avw                          8.839       0.80    8.14       50.96
14548                        -.0265     -2.50    -.0273     -2.64
w4548                        -.0016     -0.12
opbal                        0.0097      3.88    0.0099      4.20
clbal
op45                         0.0004      0.27
cl48                         0.0027      2.25    0.0030      3.49
totdep                      -0.0049     -7.30   -0.0049     -7.61
max

Number of observations         837                 837

Probability > [chi.sup.2]      0.000               0.000
Log likelihood              -430.9834           -444.8251
Pseudo [R.sup.2]               0.245               0.221

Variable                        dy/dx       Z       dy/dx       Z

sex
thud *
inner *                        .1228     2.48     .1371      2.97
outer *                        .0711     1.09     .0893      1.42
relpanic *                     .4552     8.29     .4343      8.02
whcollar *                    -.0745    -0.75   -0.0906     -0.96
ric *                        -0.2621    -2.45    -.2531     -2.25
farming *                     0.0409     0.78
prole *
spinster *                   -0.1712    -2.39   -0.2024     -3.09
minor *                      -0.0936    -1.19   -0.0903     -1.16
dateop                       -0.0001    -3.03    -.0000     -2.22
avl                         103.5        6.45    94.57       6.17
avw                          14.84       1.66    7.970       1.04
14548                         -.0308    -2.64    -.0429     -3.45
w4548
opbal                         0.0096     4.12    0.0074      3.39
clbal                         0.0041     3.14    0.0048      3.92
op45
cl48                          0.0045     3.27    0.0050      3.87
totdep                       -0.0035    -4.54
max                          -0.0064    -4.03   -0.0106     -7.74

Number of observations      Mean
                            value
Probability > [chi.sup.2]
Log likelihood
Pseudo [R.sup.2]

Variable

sex                           .441
thud*                         .377
inner*                        .429
outer*                        .154
relpanic*                     .117
whcollar*                    0.045
ric*                         0.017
farming*                      .319
prole*                       0.090
spinster*                    0.070
minor*                       0.068
dateop                      12/42
avl                          0.0026
avw                           .0015
14548                        2.194
w4548                        1.500
opbal                        20.14
clbal                        31.18
op45                         29.56
cl48                         31.15
totdep                       68.04
max                          47.12

Note * dy/dx is for discrete change of
dummy variable from 0 to 1.

Table 3 Accounts closed, 1853-1858

                             1853-1855   February-    1857-1858
                                         March 1856

No.                           199         75           191
Female (%)                    45.7        46.7         37.7
Average opening balance       17.5        20.7         19.0
  ([pound sterling])
Average closing balance       23.3        29.8         28.1
  ([pound sterling])
Average date open            Oct. 1850   Dec. 1849    Nov. 1854
Thurles address (%)           34.4        36.7         37.7
In trust [%]                  32.7        38.7         44.0
Withdrew at same time         23.6        50.7         40.3
  as another with
same surname/address (%)
Status or occupation
  where given (%)
Farming (including family)    30.1        47.9         28.2
Labourers, servants,          11.5        18.3         10.7
  dealers, etc.
Married women, widows,        26.9        20.0         27.5
  spinsters
Minors                         5.4         4.2         13.0
Gentlemen, corn dealers,       5.4         1.4          6.9

  doctors
RIC                            5.4         1.4          2.3
Other                         14.6         6.8         10.7
Total                        100.0       100.0        100.0
Not given                     69          19           59
COPYRIGHT 2009 Sage Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Grada, Cormac O.
Publication:Irish Economic and Social History
Article Type:Essay
Geographic Code:4EUIR
Date:Dec 1, 2009
Words:7504
Previous Article:'A violent society'? Homicide rates in Ireland, 1831-1850.
Next Article:The Irish in the mining industry in England and Wales: evidence from the 1881 census.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters