From Savings Banks to Roth Accounts, a Timeline of How Americans Learned to Save

Every few years a commentator announces that Americans cannot save, and every few years the commentator is wrong in the most American way possible: by assuming nobody would build a machine for it. We built several. Some were run by a few Quakers with a ledger, some by Congress with a pen, and one by a Senator from Delaware whose last name ended up on an account type.

Let me walk you through the whole sequence, with dates, because the order matters. The habit came first. The programs came later, and where they worked, they worked because they were built on a habit that already existed. That is worth knowing before the next politician tells you that ordinary people need a government scheme to be thrifty. This is a history, and it is general information, not advice about anyone's own savings.

Before the programs: thrift as a habit

There was no federal savings policy in 1790. There was no retirement account, no deposit insurance, no payroll deduction. There was the family farm, the village store, the church and the tradesman's tin box. Saving meant keeping some of what you earned out of the hands of the tavern, the tax collector and the next bad winter, and the instrument was usually a mattress or a trusted neighbour.

The difficulty was that a wage earner with a few dollars to spare had nowhere to put them. The commercial banks of the early republic served merchants and large borrowers, and they were not interested in a laborer's deposit of five dollars. A working family that wanted to save had to find somebody to hold the money, and every arrangement carried a risk.

So the next step was the one Americans have always taken when an institution is missing. They built it.

1816 to 1911: private thrift builds its own banks

1816: the first mutual savings banks

Two institutions opened in 1816, and each claims a place at the head of the line. The Philadelphia Saving Fund Society and the Provident Institution for Savings in Boston were both founded that year, modelled on earlier British and Scottish examples. They were mutual savings banks, which means they had no stockholders to enrich. They were owned, in effect, by the depositors, and the profits stayed in the pot.

Why it mattered: it was the first time a laborer, a seamstress or a clerk could walk into an institution built for deposits of a few dollars and expect to be treated as a customer. The founders were mostly civic-minded merchants and professionals who wanted the working poor to build a cushion rather than depend on charity. The idea that a person who saves a little is a person who owes less to anyone is a very old one, and it was put into practice at the corner of the street.

1831: the building and loan

By 1831, a group in Frankford, Pennsylvania, had formed what is generally regarded as the first American building and loan association, the Oxford Provident Building Association. The mechanism was beautifully simple. Members paid in small sums on a regular schedule, and the pooled money was lent, one member at a time, to buy or build a home. When a member's loan was repaid, the money was lent again.

Why it mattered: saving and home ownership were joined in a single habit. The family that kept up the payments was saving, in the sense that the house was its savings account, and borrowing, in the sense that the money arrived before the full sum was in hand. The thrift institutions that grew from this model, the savings and loans, became the principal source of home mortgages for generations.

The nineteenth century: panics and the lesson of failure

There is no single date for this entry, because the point is general. Before there was deposit insurance, bank failures were a recurring feature of American life, and when a bank failed, depositors lost what they had put in. A family could spend thirty years building a small reserve and lose it in a week. The record of the nineteenth century is full of panics, and the lesson that depositors took from them was that a bank's good name was the only protection they had. That is the context in which the next two entries should be read.

1911: postal savings

Congress authorized the Postal Savings System in 1910, and the post offices opened for deposits in 1911. Its purpose was to give people a safe place for savings: those who distrusted banks, those who lived where there were none and many recent immigrants. Deposits earned a low rate of interest, and the government stood behind them.

Why it mattered: it was the first national program for ordinary savers, and the fact that it was needed shows how thin the private network still was in much of the country. The system declined as bank deposit insurance arrived and ended in the 1960s. It is a good example of a government program that did a real job and was then wound down when the job was done, which is rarer than it ought to be.

1917 and 1918: Liberty Bonds

When the United States entered the First World War, the Treasury sold Liberty Bonds in small denominations to millions of ordinary Americans. Whatever one thinks of the war, the effect on the habit of buying a security was large. Many households owned a government bond for the first time in their lives.

Why it mattered: it taught a generation that a piece of paper could be a store of savings and that a household could be an investor. It also put the government in the position of a borrower from its own citizens, a role it has never relinquished.

1933 to 1941: Washington becomes the insurer

1933: the Banking Act and the FDIC

The Banking Act of 1933, often called Glass-Steagall after its sponsors, created the Federal Deposit Insurance Corporation. Deposit insurance began on January 1, 1934, at an initial coverage of $2,500 per depositor. The coverage limit has been raised many times since, and the present figure should be checked at FDIC.gov.

Why it mattered: it ended, for ordinary depositors, the bank run. A person no longer needed to be first in line at the teller's window to be sure of getting the money back, because the money was insured. The result, over the decades, was a banking system in which a depositor could stop thinking about the bank's solvency and start thinking about the deposit. There is a defensible argument that insurance also encouraged risk-taking by banks, and the savings-and-loan crisis of the 1980s is the standard exhibit. But the central achievement of the program, the end of the old runs on small banks, is not seriously disputed.

1935 and 1941: savings bonds

The Treasury began selling savings bonds in 1935, in small denominations aimed at the household investor, and the program expanded dramatically with the Series E bonds sold from 1941, as the country prepared for and entered the Second World War. The bonds could be bought for a fraction of their face value and held until maturity, and they were sold at post offices, banks and through payroll deductions.

Why it mattered: payroll deduction for savings is a bigger idea than it sounds. It made saving automatic, taking the decision out of the weekly argument and putting it on the pay stub. That is the same principle that makes a modern retirement plan work, and it was learned, in part, from the bond drives.

1935: Social Security

The Social Security Act of 1935 created a national retirement system funded by payroll taxes. It is not a savings account in the ordinary sense, since current taxes pay current benefits, and I will leave the long debate over its design to other posts. For the timeline, the relevant fact is that it established the expectation of a retirement beyond work. It also left a large gap, which private saving was left to fill, and the entries that follow are largely the story of how that gap came to be filled.

1974 to 1981: the tax code learns to leave you alone

1974: ERISA and the IRA

The Employee Retirement Income Security Act, signed in 1974, set federal standards for private pension plans, covering such matters as reporting, vesting and the responsibilities of those who manage the money. It also created the Individual Retirement Account, a way for workers to save for retirement with a tax advantage, including those whose employers offered no plan.

Why it mattered: this is the point at which the saving vehicle stops being an institution and becomes a tax rule. The IRA is not a bank or a fund. It is an arrangement under which income saved in a particular way is taxed differently. The government did not take your money and promise something later. It said: if you put it here, we will treat it differently. In 1981, the Economic Recovery Tax Act extended IRA eligibility to all workers with earnings, including those already covered by an employer's plan.

1978: Section 401(k)

The Revenue Act of 1978 added a provision to the tax code, section 401(k), that clarified the tax treatment of certain employer plans in which employees could choose to defer part of their pay. It was a technical provision. Its authors did not intend to create the dominant retirement vehicle of the next half-century, and the IRS did not issue its guidance until 1981, after which employers began to build plans around it.

Why it mattered: the 401(k) shifted the main burden and the main control of retirement saving to the worker. In the older pension, the employer made a promise and bore the risk. In the 401(k), the worker decides how much to contribute and how to invest, and bears the risk. There are reasonable arguments both ways about whether that is a gain, but it is hard to deny that it put ownership of a retirement account in the hands of tens of millions of people who would never have been able to own a pension fund. How these accounts work today is explained in How a 401(k), a Traditional IRA and a Roth IRA Really Work, in Twelve Questions.

1997 to 2006: after-tax saving and the default

1997: the Roth IRA

The Taxpayer Relief Act of 1997 created the Roth IRA, named for Senator William Roth of Delaware, who championed it. Accounts were available from 1998. The key difference from the traditional IRA is the timing of the tax: contributions are made with money that has already been taxed, and, if the rules are followed, qualified withdrawals in retirement are not taxed again.

Why it mattered: it gave savers a choice about when to pay the tax, now or later, and thereby gave them something that earlier law had not: a tax-free harvest. It rewarded the saver who was willing to pay in advance. The trade-off between the two kinds of account is laid out, with a worked example, in Pay the Tax Now or Pay It Later? Traditional and Roth Accounts Side by Side.

2006: the Pension Protection Act

The Pension Protection Act of 2006 encouraged employers to enroll workers in retirement plans automatically, with the right to opt out, and to invest the contributions in a default fund. It does not change the basic rules of the accounts. It changes the starting position, and starting positions, as anyone who has ever failed to cancel a subscription knows, are enormously powerful.

Why it mattered: it acknowledged that the main obstacle to saving was not ignorance but inertia. Make the saving automatic, and many more people save. Whether the default should be an opt-in or an opt-out is a question on which reasonable conservatives differ, since a default is a nudge, and some of us are wary of nudges. But the evidence that people stay in a plan they were placed in is hard to ignore.

What the timeline teaches

Let me give you the argument in order, because I think the story supports four claims.

1. The habit came first. The mutual savings banks of 1816 and the building and loans of 1831 were built by private citizens who saw a need. Nobody in Washington asked for them.
2. Government's best contributions were rules, not rival institutions. Deposit insurance made private banks safe to use, and the IRA and the 401(k) changed how private savings are taxed. In each case the government acted on the framework, and the money stayed in private hands, in private accounts, owned by the saver.
3. Automatic beats aspirational. Payroll deduction for bonds in the war years, the 401(k) and the 2006 default all worked because they removed the decision from the weekly budget.
4. The saver is the owner. The most successful programs in this timeline let ordinary people hold the title to their own money. That is a conservative idea, and the history supports it.

Now, a caution about the optimism. The record has failures too: bank failures before insurance, the savings-and-loan crisis, retirement accounts that people neglect or raid. It is a story of improvement, not perfection. And a country that builds ever more machinery for saving while its own government borrows without limit is sending a mixed message. The federal books are open for inspection at the national debt clock.

The constitutional anchor

There is a constitutional point hiding in this timeline, and it is one a lawyer might smile at. Article I, Section 8, Clause 1, opens with these words: "The Congress shall have Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States."

Notice what Congress did with that power in 1974, 1978 and 1997. It did not spend. It did not build a national savings bank and run it. It used the taxing power in reverse: it said that income saved in a certain way would be taxed less, or later, or not at all. A tax rule that leaves money with the person who earned it is the mildest use of a power the Constitution grants and a perfectly respectable one. Whether the particular rules are wise is a policy argument. That the form, deferral and exclusion for private accounts, fits the character of a government of limited powers is, I think, hard to dispute.

It also fits the older American instinct, the one Benjamin Franklin gave voice to in the 1758 collection of maxims known as The Way to Wealth, that self-reliance is both a private virtue and a public good. His sermon is read line by line in Benjamin Franklin's The Way to Wealth, Read Line by Line for Modern Savers. A people who save are a people who do not need rescuing, and a free republic runs better on that kind of people.

The case in brief

1. American saving began as a private habit: the mutual savings banks of 1816 and the building and loan of 1831 were built by citizens, not by the government.
2. Government's largest contributions were framework rules: deposit insurance under the Banking Act of 1933, and tax treatment for the IRA in 1974, the 401(k) in 1978 and the Roth IRA in 1997.
3. Programs that made saving automatic, from payroll bond deductions to the 2006 default rules, worked because inertia is stronger than intention.
4. The best of these arrangements leave ownership with the saver, which fits a government of limited powers acting through the taxing clause of Article I, Section 8.
5. The record includes failures as well as successes, and a nation that encourages thrift at home should expect its own government to practice it.

What you can do: read the first sentence of Article I, Section 8, Clause 1, then check the current limits on your own retirement accounts at IRS.gov.