Arnold Kling  

Debating Forced Saving

PRINT
Prescott on Social Security... Explaining Pharmaceutical Regu...

Tyler Cowen writes,


First, how much can our government force people to save in the first place? You can make them lock up funds in an account, but they can respond by borrowing more on their credit cards, taking out a bigger mortgage, and in general investing less in their future. The net increase in savings will be much less than the mandated increase.

In contrast, Jane Galt writes,

So if your fellow citizens are willing to provide some minimum level of subsistence for you in retirement, the temptation grows to risk bankruptcy in retirment, either by blowing all your money on wine, women and song, or by making extra-risk investments in the hope of extra-juicy returns.

Also, many people aren't good at planning. A look at the current state of boomer finances, or my 401(k), would scare the hell out of you.


My view in this case is paternalistic. I do not think people are financially sophisticated enough to know how much they should save or to know how to offset very much of forced saving.

My view is that hardly anyone understands just how much the increase in longevity and improvement in medical care that has taken place in recent decades implies a need for larger saving. In Saving Freedom, I showed in a simple example that we might need five times as much saving at age 65 as we needed 50 years ago.

Tyler continues,


Private accounts meet further problems if people live for a long time. What about the woman who survives to 105?

I think that the key will be getting her to retire at 75 rather than 65. If she retires at 75, and her savings are exhausted at 90, I do not mind government picking up the rest of the tab.

He concludes,


I stand where I started, namely that social security should evolve into a welfare system for the elderly, but without the forced savings component.

My concern is that without forced savings, the vast majority of people will require the welfare component, so we would end up with something that looks like today's tax-and-transfer scheme.

For Discussion. To what extent do you think that people would offset a forced-saving scheme with greater borrowing or other means?


Comments and Sharing


CATEGORIES: Social Security



TRACKBACKS (3 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/156
The author at The Club for Growth Blog in a related article titled Monday's Daily News writes:
    Litigious Risks to Your Health - Steve Moore, Wash Times House GOP Steers Course For Limited Gov’t - Mike Pence, HEO New Texas Lawmakers Learning the Ropes - Houston Chronicle Short List of Supreme Court Nominee - M. Kirkpatrick, WSJ... [Tracked on November 15, 2004 9:50 AM]
COMMENTS (12 to date)
lindenen writes:

So how much money would a 26 year old need to retire eventually?

Part of the problem with government picking up the tab is that this may turn her into a profligate spender with whatever savings she does have.

spencer writes:

Linden --If over the last 40 years -- since 1965 -- you had earned average hourly earnings you income would have grown from about $5,000 in 1965 to about $27,500 in 2004. If over this time you had saved 2% of your income you would have saved about $12,000.
If you had invested this in the s&p 500 and reinvested the dividend and obtained the mkt gains and loses each year you would now have a portfolio of about $112,000. If you invested this to provide a 5% yield it would give you an income equal to about 20% of you final year income in the 40th year of your working life. If you had saved 5% of your income each year and retained all the other assumptions it would give you an income equal to about 50% of you final year of working.

If you had saved 10% it would give you an income equal to your final working year.

The final years have a big impact on the results.
Sticking with the original assumptions of saving 2% of your income, in 1994 -- the 30th working year you income would equal 9% of you final year inceme. At the mkt peak in 1999 your income would be equal to 27.4% of your final year income

Does this answer your question.

Jim Erlandson writes:

Before we start any discussion about the future of Social Security, we need to define the objectives. I suggest the following:

US Citizens older than "A" years will be guaranteed an income of $X,XXX per month for life. (insert whatever numbers you like)

To achieve this objective, each individual will be required to deposit "B"% of all income into an individually owned account, managed by an insurance company with the principal and interest (a rate determined annually and guaranteed until payouts begin) to be applied to an Life Annuity with payout to begin (at the option of the owner) at age "A". Once the balance in the account is large enough to achieve the guaranteed income, the owner may (at his option) stop paying into the account.

Like with all annuities, the owner will have the option of taking a smaller payout in exchange for having the unpaid balance paid to a beneficiary or estate upon the owner’s death. If the owner dies before retirement, the balance is paid to a beneficiary or estate.

The owner will also be able to pay more than the minimum to put more money to work sooner.

Retirement benefits (the guaranteed income) for individuals who don’t pay enough into their account (chronically unemployed, disabled etc), should be covered by the government out of general funds. After all, “we” decided to make the guarantee, so “we” should be on the hook for it just like we are for other welfare programs.

With this type of “forced savings” scheme, how people conduct their financial lives (profligate spenders, borrowers, gamblers) is of no concern.

p writes:

Spencer, nice job presenting an analtical example. I must say that while the average income provides a staring point it is not an accurate yard stick. The average is not the most likely income at the end of one career (assuming you started there as well).

Your point in regards to 1999 asa final year is excellent. An insurance component is key to managing investments.

spencer writes:

P -- you are quite right -- iam thinking of doing the numbers on the assumption that you start at the averge and end with double the average --
going from $5,000 to $55,000. A more realistic example.

I started looking at the problem this way because almost all analysis about savings, retirement income, and ss almost completely ignores this issue and just works on simple asumptions of constant wages -- ie, no wage growth.

Because of this their analysis of savings is very unrealistic.

Boonton writes:
So how much money would a 26 year old need to retire eventually? Part of the problem with government picking up the tab is that this may turn her into a profligate spender with whatever savings she does have.

A very potent question, a while back on one of the other numerous comment lists on SSI I did some regressions analysis to examine whether the SSI surplus caused gov't to increase its spending. I found no such connection and some evidence of just the opposite.

One way we could test this hypothesis on consumer behavior is to ask ourselves what happens to savings as public confidence in SSI falls. If the public thinks SSI is a great system (as they did in the 60's-late 70's) then one would expect to see a fall in private savings. After all, if you trust that SSI will take care of you why would you want to save more? On the other hand, its clear many people are now starting to doubt SSI and advocates of privitaztion have convinced many that SSI simply will not be there in the future. This should result in increased savings by those groups who buy this argument the most (the young and those a bit beyond young). Has it?

Boonton writes:
US Citizens older than "A" years will be guaranteed an income of $X,XXX per month for life. (insert whatever numbers you like) To achieve this objective, each individual will be required to deposit "B"% of all income into an individually owned account, managed by an insurance company with the principal and interest (a rate determined annually and guaranteed until payouts begin) to be applied to an Life Annuity with payout to begin (at the option of the owner) at age "A". Once the balance in the account is large enough to achieve the guaranteed income, the owner may (at his option) stop paying into the account.

This works out fine until you notice that the insurance/annuity company will require a certain $ amount in order to guarantee a lifetime annuity of $X per year. For those with very high incomes, this will be no problem. What happens for those with low incomes? Even paying B% into the fund they may not build up a balance sufficient to purchase this type of annuity.

The result is you would have to either force the insurance company to provide it to them anyway (which means the system becomes an even more disguised transfer scheme since the insurance company will make up for the costs by charging more for the annuity). Add in the fact that the Fed. gov't would have to step in with tax dollars if the insurance company went belly up on bad investments (and before you say 'strict regulation' remember in good times the incentive would be for company lobbyists to demand looser rules) and you will suddenly discover you are back at the old SSI program.

spencer writes:

P -- agree with your earlier critism that need to use a different income growth assumption than one that a person earns the average income over their work lifetime. So I assumed that an individual starts with the averge income and ends
with an income double the averge income. Starts
in 1965 with an income of $5,000 and end in 2005 with an income of $55,000 rather than $27,500--
this should be a good assumption.

What this does is reinforce my conclusion that all the analysis based on an assumption of no wage growth is really a serious problem.

assume one had saves 10% of their income and invested it in the S&P 500 with the dividend reinvested at the end of the 40 years invested the portfolio to generate a 5% yield.

Under the original scenario that income rose to $27,500 this would yield an income equal to 100% of the final years income. But under the new assumption that the final years income is $55,00
the income is only about 60% of the final years income. The reason for this is that under the second scenario income growth relative to stock market growth is much stronger. In the first scenario the rise in the stock market was 2.6 times income growth but in the second scenario the market rise was only 1.17 time income growth.

Arnold Kling writes:

I think that saving 2 percent of your income is ridiculously little. If you work for 40 years and retire for 20 years, then ignoring compound interest you need to save something like one third of your after-tax income while you are working. That way, if you have $30 K after-tax income, you can spend $20K per year both while working and while retired. Compound interest reduces the required saving rate, but it does not take you from 33 percent to 2 percent.

Nigel Kearney writes:

What about the political problem, i.e.:

a) If you continue to provide for those currently retired or nearing retirement (and surely you must), then you are asking current workers to pay twice, once for others and once for themselves. Hard to sell when you need votes to win an election.

b) The existence of people's money in private accounts where they can't touch it creates the possibility of a campaign tactic such as:
"Vote for me and I'll give you your money back."
If people raise concerns about their retirement, the candidate can come back with something like:
"You paid for your kids' childhood, why shouldn't they pay for your retirement?"

This is exactly what happened here in New Zealand in 1975 and we've been wondering how to find the money to pay for government superannuation ever since.

Any thoughts on how, politically, a compulsory private account system could be put in place to beging with, and then kept safe from unprincipled politicans of the future?

Boonton writes:

Yea how about the Social Security system with its rules that are so arcane that no one really understands them? It does serve as a deterrent to forms of mischief such as 'borrowing' against your social security to buy a house or for a 'personal loan'?

Ken writes:

"b) The existence of people's money in private accounts where they can't touch it creates the possibility of a campaign tactic such as:
"Vote for me and I'll give you your money back."
If people raise concerns about their retirement, the candidate can come back with something like:
"You paid for your kids' childhood, why shouldn't they pay for your retirement?"

This is exactly what happened here in New Zealand in 1975 and we've been wondering how to find the money to pay for government superannuation ever since."

So the politician successfully argued that children should pay for their parents' retirements, then got into office, cut compulsory savings, and left the guaranteed retirement income in place?

He was doing great until that last bit.

I'm not ready to label "unprincipled" legislators that permit you to have your own money. Promising you a free lunch if you're lucky enough to live to a certain age - that's the real unprincipled legislative act, and there's been more than enough of that.

"My view is that hardly anyone understands just how much the increase in longevity and improvement in medical care that has taken place in recent decades implies a need for larger saving."

Of course it also enables a longer working life. There's no law that says people have to retire at 65, and people are much more able to figure out when they get to that age that they're not quite ready to be put out to pasture, and can't afford it anyway, and they'll keep working. Problem solved.

Comments for this entry have been closed
Return to top