The History of Saving: Savings Banks, Deposit Insurance, and the Economics of Thrift
1810: The Savings Bank Movement Begins
Before the nineteenth century, ordinary wage earners had almost nowhere to put small sums. Commercial banks served merchants and required minimum balances far beyond a labourer's reach, so household reserves took the form of cash, livestock, or membership in a mutual aid society. In 1810 the Reverend Henry Duncan founded a savings bank in the Scottish parish of Ruthwell that accepted very small deposits and paid interest on them. The model spread rapidly through Britain and then across the Atlantic: the Philadelphia Saving Fund Society and the Provident Institution for Savings in Boston both date from 1816. What these institutions changed was not the desire to save but the mechanism — for the first time, a small sum could be stored somewhere it earned a return and could not easily be spent by accident.
1933: Deposit Insurance Changes What a Deposit Is
A savings bank is only useful as a store of value if the bank survives. Between 1930 and 1933 thousands of US banks failed, and depositors who queued at a failed institution often recovered only part of their balance, or recovered it years later. The Banking Act of 1933 created the Federal Deposit Insurance Corporation, which began insuring deposits on 1 January 1934 with an initial limit of $2,500 per depositor. The limit was raised repeatedly over the following decades, reaching $100,000 in 1980 and $250,000 on a temporary basis in October 2008 before being made permanent by the Dodd-Frank Act in 2010. Federally insured credit unions received comparable share insurance through the National Credit Union Administration from 1970. Deposit insurance converted an insured bank deposit into an instrument with essentially no nominal default risk to the depositor.
1936: Keynes and the Consumption Function
John Maynard Keynes's The General Theory of Employment, Interest and Money (1936) put household saving at the centre of macroeconomics. Keynes proposed that consumption rises with income but by less than the full amount, so the fraction saved rises with income — the relationship that became the consumption function, with its slope the marginal propensity to consume. He also described the paradox of thrift: an increase in the desire to save, if it occurs across an entire economy simultaneously, reduces aggregate spending, and therefore income, and so may not raise total saving at all. The paradox is a statement about aggregation, not about any individual household, and confusing the two is one of the most common errors made when the idea is quoted outside its context.
1954-1957: Life-Cycle and Permanent Income Theories
Two closely related theories reframed saving as a problem of timing rather than of thrift. Franco Modigliani, working with Richard Brumberg, published the life-cycle hypothesis in 1954: individuals plan consumption over an expected lifetime, borrowing when young, accumulating during peak earning years, and drawing down in retirement, so that consumption is smoother than income. Milton Friedman's A Theory of the Consumption Function (1957) advanced the permanent income hypothesis, distinguishing the permanent component of income, which drives consumption, from transitory fluctuations, which are largely saved or borrowed against. Both theories predict that a one-off windfall is treated differently from a durable raise, a prediction that has been tested extensively and holds only partially in practice.
2004 Onward: Behavioural Saving and Automatic Enrolment
Richard Thaler and Shlomo Benartzi published Save More Tomorrow in the Journal of Political Economy in 2004, describing a programme in which participants commit in advance to raising their contribution rate at future pay increases. The design responds to observed behaviour rather than to theory: present-focused decision-making, inertia, and loss aversion each work against a contribution increase made today but not against one scheduled for later. Related research on default effects found that participation in workplace retirement plans rose sharply when enrolment was made automatic with an opt-out. The Pension Protection Act of 2006 in the United States gave employers a clearer legal footing for automatic enrolment and automatic escalation, and both features became widespread in the years that followed.