Arnold Kling  

Defending Social Security

PRINT
Asymmetric Price Adjustment... Trade and the WTO...

In the March 2004 issue of the the American Economic Review, Peter Diamond defends the structure of Social Security. (I cannot find the article on the web, unfortunately.) He argues that in the absence of Social Security, individuals would save too little, under-utilize annuities, and under-purchase life insurance.


To my mind, the heart of the context for thinking about Social Security is that it substitutes for poor decision making and for missing insurance opportunities...

significant numbers of workers do not insure their lives adequately...Since the government cares about the different family members (and not just the worker), direct allocations to family members matters since they will change the allocation of resources within the family. Protecting family members is a role governments have recognized for centuries.


I think that I would agree with just about all of Diamond's positive analysis (economic description) of Social Security and just about none of his normative analysis (value judgments).

For example, Diamond considers it a big deal that the market for private annuities is not well developed. As a result, my mother-in-law, who has liquid assets but no annuity, could over-spend or under-spend, depending on how long she lives. If she dies unexpectedly early, then she will leave a relatively large bequest. If she lives unexpectedly long, she could leave a small bequest or even end up depending on her children for support in her last years.

I think that many familes do not mind sharing this risk. I certainly would not have a problem with supporting my mother-in-law if she lives into her 90's. By that time, my own children will be well past college and should be on their own. Looking at my own situation, if I die early and leave a large bequest "by mistake," that will come when my children still need it, whereas if I live longer and the bequest is diminished, that will be after my children have had time to establish themselves. For me, purchasing an annuity would not necessarily be optimal relative to the multi-generational life cycle.

Furthermore, for elderly people, it oversimplifies things to think that you go directly from living in good health to sudden death. Instead, you may go from living in good health to incapacitation, or to contracting a terminal illness, such as cancer. How you wish to allocate your resources over the remainder of your life could be affected by those developments.

Leaving your wealth in a lump sum, rather than purchasing an annuity, gives you more choices in case of incapacitation or terminal illness. I think that this is a valid, rational reason--an option value, if you will--to hold onto one's savings rather than purchase an annuity.

For Discussion. How comfortable are you with Diamond's paternalistic assumption that if people do not act optimally for themselves then government should step in to correct their behavior?


Comments and Sharing


CATEGORIES: Social Security



TRACKBACKS (3 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/70
The author at Knowledge Problem in a related article titled CARNIVAL OF THE CAPITALISTS 19 APRIL writes:
    Now, your Carnival of the Capitalists ... (note: next week's host is Venturpreneur) Given that we've just survived April 15, and given that it felt to me like I'd been beaten about the head and shoulders, I thought we'd have... [Tracked on April 19, 2004 8:44 AM]
COMMENTS (81 to date)
Scott Johnson writes:

Well, many many people do not act optimally for themselves, and while I don't know that government should or can take care of them instead, I'm concerned about what happens if they don't. Wasn't poverty amongst the elderly a huge problem that social security was designed to take care of? What was it like if that was so?

Shank writes:

Seems like many people would have a hard time believing that the government would not assist them in their old age if they were destitute. I think they'd probably be right. If a lot of elderly people made bad financial decisions in their younger years, society may be crawling with hungry, homeless, elderly people. What politician could support letting Grandma Betty starve to death?

Therefore, what incentive would people have to make prudent financial decisions in their working years? If we no longer have social security, I would make sure to save enough for my own retirement. The larger paycheck would not benefit me now, as the increased amount would be put in a long term savings account or guaranteed interest or conservative investment. However, when poverty is high among not-so-prudent elderly people, my taxes will go up to cover food/housing/welfare payments for them. As a prudent person, my situation will worsen

I don't like paternalism per se, but when it protects the prudent like me from being drained by the spendthrifts, then I think it is acceptable.

Rob writes:

I certainly have a problem with Diamond assuming that if my decisions don’t match to what he thinks is optimal for me, then the government should step in and force the decision upon me. Because of the myriad of factors that go into determining my preferences, my optimal decisions cannot be quantified. Therefore, we should actually define the optimal decision as the decision the individual makes. As I understand it, this is what most economists do. When an economist’s theory does not match the decisions made by individuals, we look to correct the theory, but you do not assume that the individual is wrong, requiring behavioral correction.

We do, however, have to realize that there are some people who are risk adverse and others who love risk. The risk takers will be more likely to end up destitute in their golden years than the risk adverse. Now some, like the previous commenter, will state that we do not want to live in a society that will allow those risk takers to live up to the full consequences of their actions; we do not want to see Grandma Betty starve to death. Since we all know that we will require a safety net from the government, the risk adverse will assume more risk than they would otherwise, leading to more people depending upon the safety net.

That said, there are several options that could still be better than the current Social Security system. For example, perhaps the government could require certificates of minimum retirement insurance from all laborers in much the same way we require drivers to be insured. To go through all of these options, however, would be to diverge from the point of discussion, which is Diamond's paternalistic assumption.

JD writes:

Social Security was NOT initiated or conceived to be a replacement for retirement savings or insurance planning, except for the poorest sector of Americans. However, it did not take long for society to embrace it as a handout and demand more and more. If anything, Social Security provided a discentive to save. Diamond conveniently understates this; the State is empowered by the impact of Social Security, creating a larger and larger dependent class precisely because responsibility is shifted from the individual to the State. What happens if the dependent class becomes a sizeable enough to control the State (i.e. through elections)? What happens if they want to vote themselves a pay raise, or vote against a pay cut? We'll know in the next 15-20 years.

Gordon writes:

How did people in this country survive for 300+ years before SS? SS was a political gimmick. The retirement age was set so that, on the average, benefits would be paid for only a year or two. Government is only a group of people who think they know what is best for everyone else. Did you ever wonder why Congress excluded itself from SS, only to create their own sweetheart retirement plan?

Xmas writes:
How comfortable are you with Diamond's paternalistic assumption that if people do not act optimally for themselves then government should step in to correct their behavior?

That's a tough one. First, I think your question is worded incorrectly, "optimal" does not equal "adequate". I'm comfortable with the government being paternalistic enough to force adequate behaviour on its citizens. Although, I'll often disagree with how the behaviour is enforced, I wouldn't debate the need for the behaviour enforcement.

"Optimal" is a much more loaded word. I cannot think of any government rule or service that could use the adjective optimal. Optimal would necessarily imply different results for different people, an agility which is beyond the ken of any part of government.

Boonton writes:

SSI has not really created such a 'dependent class'. The only really abusive increase in benefits came from Richard Nixon and that wasn't due to seniors making unreasonable demands but from his pathological need to win reelection by 100% of the vote.

As for paternalism, I'll ask a question; Isn't the tax subsidized status of 401K's, IRA's and Roth IRA's also gov't paternalism? We could just as easily ask why those who wish to consume more today must pay higher taxes in order to reward those who lock their money away for 30 years? How about those who wish to use unconventional methods of savings such as collecting comic books? Why should their economic choices be punished by gov't policy that supposedly knows better?

Of course, most people who carp about paternalism are quite enthusiastic about 401K's. Yet if the world was as rational as they claim, then they too are supporting a form of gov't paternalism.

Eric Krieg writes:

A paradox: those who don't take investment risk in their retirement planning are actually taking a risk that they will never be able to afford retirement.

Seriously, the 50 something guy who saved dutifully into a money market account, and thus missed the stock boom, has taken a big risk. It wasn't an investment risk, but it was a risk nonetheless. The money market saver is never going to be able to retire.

Eric Krieg writes:

There is a huge difference between encouraging retirement savings through subsidy and enforcing retirement savings through a mandatory program.

If you work, you must enroll in Social Security. There are pretty much no exceptions (the Post Office is the only organization I know of that is not a part of SS).

Force is paternalistic, encouragement is not.

Eric Krieg writes:

The problem with an annuity is that it creates taxable income. There could be some situations where people make the optimal decision based on minimizing their taxes where they do not get the annuity.

The annuity also decreases your income flexibility. Most retirees consume more in their earlier years and less in their later years. The annuity is structured that way.

Finally, the annuity has risks of its own, namely the risk of early death.

I don't think that it is "suboptimal" for someone to live off his nest egg rather than buying an annuity. If it were myself, I would probably hedge my bets by buying an annuity with only a portion of my nest egg. That way I have some guaranteed income, but I also have the flexibility that the nest egg allows.

Boonton writes:
Seriously, the 50 something guy who saved dutifully into a money market account, and thus missed the stock boom, has taken a big risk. It wasn't an investment risk, but it was a risk nonetheless. The money market saver is never going to be able to retire.

The counter argument is that this is not risk. Risk is the expected variation of your return. The person putting money into the money market knews pretty much what his return was going to be. Hence he should have known that his retirement would have to be of a smaller scope.

There is a huge difference between encouraging retirement savings through subsidy and enforcing retirement savings through a mandatory program.

For the person doing the saving there is a difference but not for those who with to do other options. If, for whatever reason, I don't want to put my money into a 401K or IRA I still am required to pay taxes at a higher rate. Unlike SSI, I'm being required to 'save' for someone else's benefit which is even worse than a mandatory 'saving' that I could eventually get back in some form!

Sorry, if you want to rail against paternalism then you are going to have to take on the Middle Classes 'Newer Deal' of tax breaks.

Eric Krieg writes:

>>The counter argument is that this is not risk. Risk is the expected variation of your return. The person putting money into the money market knews pretty much what his return was going to be.

Not true. Many of these money market savers began saving in the late 1970s, when money markets paid out attractive rates of interest (and the stock market was in the doldrums).

Over the last 25 years, money market returns have tanked. If you planned on getting a 14% return on your money market, you are in for a rude awakening.

You simply HAVE to hedge your bets against not getting ENOUGH of a return by having a portion of your money in stocks.

This was the subject of an article in the Journal a couple of months ago. It was one of those "tune up your portfolio" deals where investment advisors look over somebody's portfolio and critique it. The guy had been saving since 1980, had a very good savings rate, but it was all in safe, low yield stuff. He was never going to be able to retire at the rate he was going, and he was only 50.

Eric Krieg writes:

>>If, for whatever reason, I don't want to put my money into a 401K or IRA I still am required to pay taxes at a higher rate. Unlike SSI, I'm being required to 'save' for someone else's benefit which is even worse than a mandatory 'saving' that I could eventually get back in some form!

Just because an individual is paying a lower rate doesn't imply that you are subsidizing that individual. That is a static analysis, it doesn't take into account the "unintended consequences" of people changing their behavior due to new incentives.

For example, 401(k)s are only taxed deferred, not tax free. Taxes will eventually be paid on that money. It could easily be the case that tax rates in the future will be higher than tax rates in the past, and MORE taxes will be paid on 401(k) money than if the money had gone into a taxable account.

Boonton writes:

That wouldn't matter Eric, if taxes were lowered accross the board in exchange for abolishing 'incentives' then it could very well be that 30 years from now income would be higher but so would tax rates....

Spending via the tax code remains essentially spending. There's little real difference if the gov't gives me a $0.25 check every time I put a $1 into a 'preferred account' or if the gov't lowers my taxes by $0.25. In both cases the 'incentive' may be a beneficial policy but it is silly to pretend its a policy that comes without any costs at all. If the policy does have costs then someone must be bearing them, and who could be bearing the burden of 401K policy? It must be people who, for whatever reason, choose not to use 401K's!

So we are back where we began, if we are serious about rejecting paternalism then 401K's should go as well. Why should a low saver be deemed less rational than a high saver? Why would the market be unable to generate the 'correct' amount of savings when we assume it generates the 'correct' amount of prices and quantities for almost every other product in the economy?

Joe Glandorf writes:

Suppose the government could devise a truly "adequate" insurance and retirement formula. $100,000 for death, $20,000 per year of retirement income for example.

How long would it be before very sad stories appeared in the paper of people who "lost their homes" due to the inadequacy of this amount for their particular circumstances? How long would it be before hearings would be held with "victims" of this "inadequate" program "demanding justice"?
How long would it be before it was "changed"? How long would it be before the next cycle ensued?

It is utterly impossible to devise a "just" or "adequate" plan that cannot be shown to disadvantage someone. One "good" victim and the inertia to begin tinkering with the numbers would be overcome and the inevitable expansion of the bureaucracy would begin.

People who don't understand this process probably have never worked in any kind of organization. The only reason bureaucracies are held in check in private business is the need for profits. The only reason they are held in check in "non-profits" are limitations on their ability to generate donations. Give a bureaucracy the ability to expand and it will until the funds run out.

So, a magical "adequate" level of "security" will never exist for long. It will always increase given political pressures. Knowing this does it make sense to even consider pursuing this course since the eventual outcome is totally foreseeable (communism)?

It most certainly is sad that individuals sometimes suffer catastrophic financial loss. However, no a tiny trade-off exists to significantly mitigate this problem without putting a horrific scenario in place, eventual totalitarianism.

Boonton writes:

Interestingly that story just doesn't fit SSI. The only time the gov't 'went wild' with the trust fund was when Nixon raised benefits for no particularly good reason.

Eric Krieg writes:

B, you are arguing semantics. Theoretically, in a static environment, the tax code is a zero sum game. But I think that the latest stats from the OMB, showing how the top quintile of income earners pay a higher share of all taxes now than in the '80s, even though tax rates now are actually lower, shows that zero sum has zero value.

Boonton writes:

So, Eric, is it your position that subsidies through the tax code have zero cost? IF that is the case then the more subsidies that litter the tax code the better. After all, if they have no cost then even if they only benefit mohair farmers it's gotta be a good thing!

Tax payments versus rates have little to do with this discussion. Regardless of income, the 401K system benefits the taxpayer who opts to use 401K's while it punishes the taxpayer who, for whatever reason, does not opt to use 401K's. This is easy to see, take two taxpayers with identical incomes and compare their taxes assuming one puts money into a 401K while the other doesn't.

You are attempting to mount a supply side argument in order to try to show the 401K system is not paternalistic in its own way. You have a difficult emperical case to make. For 2003 the top tax rate is 35% while the lowest is 10%. A dollar put into a 401K, for a taxpayer at the top bracket, will have to eventually, somehow, someway, generate $2.86 in taxable income to be a breakeven proposition for the Treasury (BTW, this is in Present Values, if you look into the future the taxable income generated would have to be much more to also account for the time value of money).

The supply side argument is even more difficult to make when one also realizes this must be income that would not have otherwize been generated! The other sad part of using the tax code as a means to subsidize behavior is the fact that a portion of the money is wasted because some of the subsidy would go to people who would have otherwise engaged in that behavior.

What does this mean? Well say in order to encourage charitible contributions they are given a 10% tax credit (you can cut your taxes by 10% of your donations). Say before this policy Americans donated $100B per year and after this policy they donated $125B. A 25% increase! Wow!... But that increase cost $12.5B (10% of $125B) because the deduction applies both to donations that would have been made anyway as well as new ones.

Well now certainly some people saving for retirement in their 401Ks would have been saving for retirement regardless of the tax code. This means that even if $1 in 401K contributions produced $2.86 in present taxable income, a portion of those 401K contributions would have been made even if 401K's had no special tax treatment. Now you must show that 401K's somehow had returns so large that this effect is cancelled out.

Once you start making this argument you run up against a 3rd barrier, the power of the market. If 401K returns were so powerful for the economy, then why is the market unable to properly reward those who contribute to their 401K's? This is the same problem I have with market oriented economists who claim Americans are not saving enough and therefore need gov't help. IF the economy is suffering a shortage of savings, why is the market unable to properly reward those who produce savings....just like it is able to properly reward those who make Beanie Babies, Pokeman cards, zippers and everything else?

Long and short, the shelfish taxpayer who does not want to use a 401K is not benefited by the 401K using taxpayer somehow producing so much extra taxable income that his taxes are either stablized or cut. Try as you might, the fact remains that 401K's are no less paternalistic than Social Security.

fling93 writes:

Just a couple off-point corrections.

Eric: For example, 401(k)s are only taxed deferred, not tax free.

Actually, the contributions are also tax-deductible. Income is taken off your W2, so you don't even have to itemize (and indeed, it can help you get below the income caps for IRAs).

Eric: It could easily be the case that tax rates in the future will be higher than tax rates in the past, and MORE taxes will be paid on 401(k) money than if the money had gone into a taxable account.

Possible, but not likely. Especially for people who are no longer drawing an income by that time. And that the future tax has to be high enough to counteract the extra returns earned by the amount that would have been taxed.

Robert Schwartz writes:
How comfortable are you with Diamond's paternalistic assumption that if people do not act optimally for themselves then government should step in to correct their behavior?

The question in my mind is how comfortable is Diamond with it? Does he think that they should be allowed to breed? or vote? If they are smart enough to vote, why aren't they smart enough to decide how to use their money?

One morning Shaw Livermore asked us: "How can you have self-government if you cannot govern your selfs?"

I am still working on that one.

Eric Krieg writes:

>>Eric: For example, 401(k)s are only taxed deferred, not tax free.

>>Actually, the contributions are also tax-deductible. Income is taken off your W2, so you don't even have to itemize (and indeed, it can help you get below the income caps for IRAs).

What other mechanism is there to defer taxes? Yeah, your 401(k) contribution comes off the top of your taxable income, just like any deduction does. It doesn't change the fact that taxes are only deferred, not avoided.

>>Eric: It could easily be the case that tax rates in the future will be higher than tax rates in the past, and MORE taxes will be paid on 401(k) money than if the money had gone into a taxable account.

>>Possible, but not likely. Especially for people who are no longer drawing an income by that time. And that the future tax has to be high enough to counteract the extra returns earned by the amount that would have been taxed.

I can create Excel spreadsheet scenarios where I myself could have more income in retirement than I have right now, making some pretty reasonable assumptions regarding investment returns, inflation, and income growth. People with modest incomes, high savings rates, and aggressive investments (fully invested in stocks) could very easily have more income in retirement than they make now.

And this is all predicated on current tax rates. All bets are off if tax rates go up, which there is a high probability of happening.

This is why the Roth IRA is so important. You hedge against having TOO MUCH taxable income in retirement by saving a portion of your investments there.

Of course, you now have the risk that some future Congress will make the Roth IRA taxable. I guess you can't hedge against every risk.

Eric Krieg writes:

Boonton, the whole idea behind 401(k)s is that taxing of savings is double taxation. In a perfect world, savings wouldn't be taxed. But since that isn't possible, and since double taxation of savings leads to people saving less for their retirement, Congress has set up the 401(k).

I could agree with your paternalism argument if there was any link whatsoever between taxation and government spending. There isn't, and thus taxation is not a zero sum game, and giving people a break on their 401(k)s is not a subsidy from one person to another. It isn't at all clear from history that if 401(k)s were eliminated, overall tax rates would go down, spending would change, or that revenues collected by the government would go up. Thus, I don't think that you can call it a subsidy.

Eric Krieg writes:

>>And that the future tax has to be high enough to counteract the extra returns earned by the amount that would have been taxed.

If you are in the same tax bracket in retirement as you are while working, there is no advantage to the 401(k). The fact that your earnings worked for you rather than being taxes just means that you have more income to tax later.

You can prove it to yourself with an Excel spreadsheet. If you have an investment that makes 5% in the first time period and 10% in the second, and the very same investment making 10% in the first period and 5% in the second, you end up with the same amount of money in the final time period.

All that taxes are is another investment expense. They lower your return when you pay them. It doesn't matter when the taxes are paid, only the tax rate matters. The 401(k) makes sense because you think that you will be in a lower tax bracket when you retire. This may not be a very good assumption.

Boonton writes:
What other mechanism is there to defer taxes? Yeah, your 401(k) contribution comes off the top of your taxable income, just like any deduction does. It doesn't change the fact that taxes are only deferred, not avoided.

The Roth IRA let's you avoid taxes on income earned from your investments. Deferred taxation is not as good as non-taxation but it nevertheless is a good. Even if the eventual tax liability is exactly the same being allowed to defer a payable 30 years is a real thing of value.

I can create Excel spreadsheet scenarios where I myself could have more income in retirement than I have right now, making some pretty reasonable assumptions regarding investment returns, inflation, and income growth.

It's certainly possible that you will be earning more when you are 75 than at any point before (maybe you'll win the lottery and be getting $100K payments every year then!). For most people the tax liability will be less; you should consider that 401K's have a lot of flexibility in distributions so people can plan their withdrawls for periods where their income will be minimal.

Boonton, the whole idea behind 401(k)s is that taxing of savings is double taxation. In a perfect world, savings wouldn't be taxed. But since that isn't possible, and since double taxation of savings leads to people saving less for their retirement, Congress has set up the 401(k).

1. It isn't double taxation. If I save $100 in 2000 and withdraw $100 from my savings account in 2005 there is no additional tax due. If in 2005 I receive $10 in interest the taxation only applies to the $10 in new income...not the $100 principle.

2. Savings is not taxed, income is taxed. What you are trying to say is that income from wages should be taxed but in a perfect world income from savings won't be taxed. Why? Why is $100 earned in a savings account or the stock market better than $100 earned by working some overtime?

I could agree with your paternalism argument if there was any link whatsoever between taxation and government spending. There isn't, and thus taxation is not a zero sum game, and giving people a break on their 401(k)s is not a subsidy from one person to another.

Let's imagine Congress enacted a law that said the tax brackets for people named Eric would be 5 points less than the brackets for everyone else. A few years later the IRS reports that total taxes paid by people named Eric has increased dramatically (since many people will change their name to Eric). Would you seriously expect us to believe that this wasn't a subsidy to people named Eric because taxes are not a zero-sum game?

Boonton writes:
You can prove it to yourself with an Excel spreadsheet. If you have an investment that makes 5% in the first time period and 10% in the second, and the very same investment making 10% in the first period and 5% in the second, you end up with the same amount of money in the final time period.

Very close Eric, I was almost fooled as well. I simulated two accounts, the first started with $1000 and grew at 5% for 25 years ending at $3,225.10. If the tax bracket was 30%, that would leave $2,257.57. That's your 401K simulation.

Then I simulated an account that started at $700 (which would be $1000 minus 30% taxes paid at the start). It grew to $2,257.57...just like where the 401K simulation ended.

BUT what you missed, Eric, is that each year you would pay taxes on that income. For example, after the first year your balance increases from $700 to $735. You had $35 in interest income which would generate $10.50 in tax (@ 30%). So your balance for year two would be $724.50...not $735.00. At the end of the day your balance will grow to $1,723.39. Compare this to the $2,257.57 you would have from the 401K type account after paying taxes on your withdrawl.

Deferred taxes do allow you to earn more income even if your tax bracket never changes. If you wish, I can send you the spreadsheet...email your address to me Brian a Ideas4now dot com.

Eric Krieg writes:

>>What you are trying to say is that income from wages should be taxed but in a perfect world income from savings won't be taxed. Why?

Because your return on investment is a reflection of the profits earned by the business you have invested in. Those profits have already been taxed at the corporate level. Thus, it is double taxation.

You can correct this in one of two ways. Either get rid of the corporate tax, or don't treat investment returns as income.

We have gone partway down this road by getting rid of the tax on dividends.

Eric Krieg writes:

>>you should consider that 401K's have a lot of flexibility in distributions so people can plan their withdrawls for periods where their income will be minimal.

There is some flexibility, but not total flexibility. There is a set age that you must start taking distributions, and there is a minimum distribution. Both rules are there so that SOME income tax is generated, and on large accounts the minimum distributions are high enough to generate significant tax bills. This is why I say that, if you are at a lower income during your working years but have accumulated a significant nest egg, you COULD be in a higher tax bracket.

I agree that the minority of people will be in this situation. My concern is that you are penalizing the people who have behaved the best, the ones who followed the rules and saved for their retirement as much as they could.

Boonton writes:
Because your return on investment is a reflection of the profits earned by the business you have invested in. Those profits have already been taxed at the corporate level. Thus, it is double taxation. You can correct this in one of two ways. Either get rid of the corporate tax, or don't treat investment returns as income.

Corporations are seperate individuals under the law. The income a corporation earns is not your income and vice versa. If you go down this road you can make make it look like any type of income has been 'double taxed'. I work for a corporation, it sells products to consumers. The consumers have already been taxed on the money they use to buy those products, hence the corporation is double taxed, hence my wages are triple taxed, hence the companies I buy from are quardruple taxed and so on!

Boonton writes:
This is why I say that, if you are at a lower income during your working years but have accumulated a significant nest egg, you COULD be in a higher tax bracket. I agree that the minority of people will be in this situation

I think it is nearly impossible to end up paying more taxes in a tax deferred account than you would have had to pay in a taxable account....even if a lottery win late in life (or getting hired by Trump to be his apprenctice) put you in the top bracket at 75 while you've been working min. wage for your whole life.

Eric Krieg writes:

>>Corporations are seperate individuals under the law. The income a corporation earns is not your income and vice versa.

The law is irrelevant. The fact is that the stockholder owns his piece of the corporation, and the ROI is his share of the corporation's profit. It is double taxation.

This is why corporations generally have not distributed profits as dividends in the past. Capital gains have been taxed less than dividends, so corporations have played the tax advantage by trying to generate capital gains rather than dividends.

This is a revealed preference, and indicates that, to corporations and stockholders at least, taxing savings is double taxation.

Income taxation of Socialist Insecurity benefits is triple taxation. You pay the payroll tax (1), but the payroll tax is not deductible, so you pay the income tax on the payroll tax (2), and then when you finally recieve benefits when you are retired, you pay taxes on that benefit (3).

Eric Krieg writes:

Boonton, do a spreadsheet. Start with a 25 year old, who starts out by making $25,000. He saves 10% of his salary in a 401(k). He works to 67. He wants to build a nest egg where he can live off the interest of the nest egg only.

What interest rate would he need to obtain in order to build a sufficient nest egg and generate $25,000 per year it?

My spreadsheet says less than 6%.

I am not taking into account either inflation nor salary increases. They will offset somewhat.

It is not impossible, or even near impossible, to have more income in retirement than you do during your working years. It may be unlikely, but even an unlikely scenario could entrap millions of people.

Boonton writes:
The law is irrelevant. The fact is that the stockholder owns his piece of the corporation, and the ROI is his share of the corporation's profit. It is double taxation.

On the contrary, if the law was irrelevant most corporations would simply reorganize themselves as partnerships where income would only be taxed once at the individual level. The owner of the corporation enjoys liability protection from this personship of his corporation. The price of the benefits of incorporation is 'double' taxation.

Boonton writes:

If interest rates are constant at 6%, a person will need to have $416,666.66 in an account in order to generate $25,000 per year forever.

My rough estimate indicates that if a person made $25K per year for 25 years and put 10% of that per year into a 401K type account earning 6%....at the end of 25 years he would have $137,161.30. If he put 25% of his income ($6,250) per year into such an account it would be $342,903.20 after 25 years.

That would yield him $20,574 per year after 25 years in interest income. Not bad but he still would have a lower taxable income than he did during his working years.

fling93 writes:

Eric: I can create Excel spreadsheet scenarios where I myself could have more income in retirement than I have right now, making some pretty reasonable assumptions

You could, but this would be the minority case, and even in that case the tax-deferral would be one of the reasons your retirement income was higher in the first place, because your capital gains and dividends can be reinvested immediately, tax-free. There is no tax drag on your investment growth, as Boonton has also said.

It's a rarer case where somebody would be better off not participating in a 401(k) at all in the long term, even assuming future higher taxes. Most likely, it involves intentionally having a higher withdrawal rate than is most tax-efficient because you have so much money you don't care (where the tax-deferral is one of the reasons you got there).

You can still be bettter off paying more money later than less money now.

Eric: All bets are off if tax rates go up, which there is a high probability of happening.

Perhaps they'll go up, but it's hardly a given they'll go up nearly enough to make it 401(k) participation a disadvantage. Especially if you consider who has the most power to influence our government and where their interest lies.

Boonton: If interest rates are constant at 6%, a person will need to have $416,666.66 in an account in order to generate $25,000 per year forever.

Also, in retirement you're better off shifting from stocks to more stable and income-generating investments and shouldn't bet on a return more than 4%.

fling93: Actually, the contributions are also tax-deductible.

Eric: What other mechanism is there to defer taxes? Yeah, your 401(k) contribution comes off the top of your taxable income, just like any deduction does. It doesn't change the fact that taxes are only deferred, not avoided.

What is deferred is not the deduction. Funds within a 401(k) are just like in a traditional IRA. The capital gains, interest, and dividends earned within the account are tax-deferred. That is, you don't pay taxes on the gains until you withdraw.

The contribution deduction lowers your taxable income. Period. You don't ever pay the "owed" taxes later. Just like any other deduction, like the mortgage interest deduction. You don't have to tally up how much taxes you saved and then pay them when you retire. There is no deferral at all there.

So yes, that, in and of itself, constitutes a subsidy, and thus paternalism.

The argument to make is that it's considerably less paternalism than Social Security, since the amount of money you are forced to pay is much less.

fling93 writes:

fling93: The contribution deduction lowers your taxable income. Period. You don't ever pay the "owed" taxes later.

Whoops. I'm wrong about that. You're right, it's deferred.

Eric Krieg writes:

Boonton, just curious, why 25 years? A 25 year old can't retire before 59-1/2, right? At least, he can't retire earlier and tap into the 401(k).

I picked 67 because that's when today's 25 year old can start getting Social Security (supposedly). I think that today's 25 year old is going to be lucky to retire at 67. God knows that they won't be retiring at 50!

My point is that, even for low earners, all that is needed is a modest savings rate, a modest ROI, and a long working life, and you could very easily have a higher income in retirement than while you were working. I started at 25, I save way more than 10%, I make way more than $25K, and I have average gains way more than 6%. I have a feeling that I am going to get screwed on the 401(k).

I should just go out and get that new Lexus RX300 with the Dub Dubs. Screw saving, its for suckers.

Funny thing, today's 25 year old will be 61 in 2038. That's when SS is supposed to go bankrupt, right?

Boonton writes:
The argument to make is that it's considerably less paternalism than Social Security, since the amount of money you are forced to pay is much less.

This is the problem with tax based subsidies, their true cost is difficult to figure out. As Eric points out, 401K's are a benefit to those who use them...hence they are a cost to non-401K taxpayers.

Boonton, just curious, why 25 years? A 25 year old can't retire before 59-1/2, right? At least, he can't retire earlier and tap into the 401(k).

No particular reason, maybe the person didn't get serious about retirement until his low 40's. The exercises should work if you use 30, 35 or 40 years.

fling93 writes:

BTW, you can withdraw earlier than 59 1/2 without penalty if you take equal and recurring distributions throughout the rest of your life. Which you'd do if you retired early. There's no reason to wait till your post-retirement income would greatly exceed your pre-retirement income.

Boonton writes:
My point is that, even for low earners, all that is needed is a modest savings rate, a modest ROI, and a long working life, and you could very easily have a higher income in retirement than while you were working. I started at 25, I save way more than 10%, I make way more than $25K, and I have average gains way more than 6%. I have a feeling that I am going to get screwed on the 401(k).

So let me do this model with the following assumptions:

1. At 25 years old, you save 15% of your income in a 401K. You earn 5% per year on your balance.

2. Tax rate is a flat 25% (to make things simple).

3. Each year your income goes up by 5%.

4. You stop working after 40 years.


At the end of year 40 your salary is $167,618.78. You have paid $452,999.15 in taxes and your 401K balance is $1,005,712.67. Pretty good.

Now if you didn't contribute to your 401K you would have paid taxes totalling $754,998.59. So over your lifetime your 401K contributions saved you taxes totalling $301,999.44. Assuming a flat 25% tax rate you can see your withdrawls from the 401K still make you a winner. (progressive taxation would make this savings even more powerful since your tax bracket would be higher as your income went over $100K). What if the tax rate is raised to 50% after retirement?

Even in that case, you will pay $502,856.34 & have the same left in your 401K but your contributions to the 401K were only $452,999.15. You still have earned $49K over and above your income. So what do you have for your savings?

1. You've netted an additional $49K of income to the person who did not save.
2. Your total lifetime consumption would:
$2,378,245.56 during your working years
$ 502,856.34 when you use your 401K
$2,881,101.89 total lifetime consumption.

A person who didn't go for the 401K would have a total lifetime consumption of:
$2,264,995.77

It would seem that even if the gov't taxed your 401K away 100% the person contributing to it got away with slightly more consumption than the person who didn't. The person who failed to contribute has over $617K less consumption to enjoy than the person who did contribute only to see himself in a tax bracket two times his working age bracket when he retires!

Eric Krieg writes:

If the 25 year old somehow gets a return of 10%, he can retire at 50 and his nest egg will generate more than $25k.

Which shows how important a % or two can make over long periods of time! Investors can't afford the high fees that investment advisors and mutual fund companies are charging.

Eric Krieg writes:

Boonton, your spreadsheet is extremely sensitive to the rate of pay increases. You get the numbers that you get because you assume a 5% rate of pay increases. Pay increases are like a ROI all of their own. They compound just like interest does.

In any case, it is clear that the reason that the 401(k) works is that investment returns are allowed to compound tax free. It isn't the fact that you don't pay taxes on the income up front that gives you the advantage. Thus, it is not a better or worse vehicle that the Roth IRA, and you can hedge your bets on tax rates by using a mix of the 401(k) (taxes paid on the back end) and the Roth IRA (taxes paid on the front end). I guess you would choose your allocation based on your perception of tax risk.

Boonton writes:

I redid the spreadsheet with 0% pay increases. In this case you would earn $1M in income over your lifetime, your 401K would have a balance of $452,999.

The contributor's lifetime consumption:
$787,500 ($1M income less $212,500 in taxes)
$226,499 (401K less 50% tax)
$1,013,999 Total lifetime consumption.

The non-contributors lifetime consumption:
$750,000 ($1M less $250,000 in taxes)
$0 No 401K
$750,000 Total consumption.

Again we find even with 0% pay raises you are still better off even if your 401K was taxed away at a 100% rate!

So the debate about 401K vs. Roth IRA is still open but the debate about 401K versus no 401K is clear. The person using the 401K is certainly receiving a benefit no matter what future tax policy turns out to be! As far as paternalism goes, how can the gov't create a benefit out of thin air? If the 401K user is getting a benefit who is paying for it? If this benefit was indeed created out of thin air by the gov't, then why was the market unable to produce this easy way to boost consumption? If the benefit isn't being created out of thin air then someone is paying for it & the better the deal is for the 401K user the worse the deal is for the person who is paying for it!

Boonton writes:

Roth IRA vs 401K

So I simulated a contribution each year to a Roth IRA of $2,812.50. This is 15% of the salary that a person who makes $25K a year less the 25% income tax from our simulation. After 40 years:

Roth Balance: $339,749.37
401K Balanc : $452,999.15

401K Advantage: $113,249.78

However, if the 401K is taxed by 25% or more the advantage is gone. So it would seem that Eric's point is accurate. The Roth IRA works if you expect your retirement tax bracket to be equal to or higher than your working one. The 401K works better, though, if you expect your retirement tax bracket to be less than your working one.

Compounding is very powerful and putting an extra $1000 away in year 1 makes a large difference in year 40. Since the 401K gives you some power over when to start taking withdrawls, most people should be able to arrange a few years where their bracket will be low. I guess for someone like Madonna who will probably be getting high income from royalities long after she stops working the Roth IRA is the best bet....

Boonton writes:

Eric's posts have lead me to an interesting observation regarding the SSI vs. 401K argument. Calls for abolishing SSI on the grounds that 401K's have better returns do not take into account taxes.

Consider the $25K per year model. He contributed $150,000 to his 401K over his lifetime and his ending balance was $212,500. That's a profitable return of $62,500. However, if the gov't puts a 50% tax on 401K withdrawls then he will only have $106,250...a loss of $43,750!

Of course, he was still better off in the 401K because we have seen he would have higher consumption even if his 401K withdrawls was taxed at 100% but looking at just his return would indicate a negative rate!

What will taxes be 30 years from now? Well we don't know but a rule of thumb is that the gov't will almost certainly not default on its debt. That means the money Bush is borrowing now (& plans to borrow over the next 5-15 years) will be taxed away from you tomorrow with interest! Advocates of privitization have been unintentionally deceptive by not taking taxation into account, it would appear!

Boonton writes:

Ok, I'm wrong!. When I figured a person's consumption during their working years I simply took taxes away from their income. I should have taken their 401K contribution as well!

A person who makes 25K a year for 40 years & contributes:

Income: $1,000,000.00
Taxes: $ 212,500.00
401K cont:$ 150,000.00

Consump: $ 637,500.00


And a person who doesn't contribute:

Income: $1,000,000.00
Taxes: $ 250,000.00


Consump: $ 750,000.00

The non-contributor has an advantage of $112,500 in lifetime consumption at retirement. However, the contributor has a 401K with a balance of $452,999. The non-contributor gets to win if the tax rate faced by the contributor is 75% or more on his 401K. If the contributor is taxed less than that rate (but still more than 25%, his working age rate) the contributor gets more consumption!

So, Eric, if you think your retirement tax bracket is going to be very...very high. On the order of 3 times your present bracket or more then you may get screwed on your 401K. However, if you assume that brackets stay where they are now there is no way you can construct a 401K where you earn a positive return but end up losing against the non-saver.

Eric Krieg writes:

Boonton, don't forget that SSI benefits are taxable. Also, SS payroll taxes are not deductible. So you would have to factor that into your analysis.

Boonton writes:

Not really since I'm not trying to estimate a return on SSI. I think the exercise is misguided but I agree that should be considered.... At least you can feel confident that your 401K contributions are a good choice unless the gov't enacts 75% tax rates on withdrawls in your future.

Eric Krieg writes:

>>At least you can feel confident that your 401K contributions are a good choice unless the gov't enacts 75% tax rates on withdrawls in your future.

Well, what's the unfunded liability of the retirement entitlements? I have seen anywhere from 5 trillion to 60 trillion dollars as estimates. The upper end is absurd, but would be something like a 50% payroll tax and a 75% income tax.

Boonton writes:

Real GDP in 2003 was about $10T. If you had a 75% tax that would be $7.5T (ignoring the supply side problems with such a tax) per year. If the unfunded liability was $60T then it would be paid off in just 8 years (60/7.5).

Of course, going into the future GDP will be higher than $10T because of economic growth & that unfunded liability is stretched out over a very long period. Somehow I don't think 75% tax rates are very likely & that's good news for your 401K.

There is a good argument that 401K's will not preserve their as much of their extreme favored tax treatment into the future. It's hard to imagine telling seniors who are poor that their SSI benefits have to be cut while those who became 401K millionaires will be allowed to live it up.

Also, the stock market is erratic so people retiring in good years will have dramatically better 401K's than those who retire in slumps. I know good money management would insist that people hedge their 401K's against risk as they get closer to retirement but will people be that rational? Look at how many of the 'best and the brightest' at Enron had 100% of their 401K in a single stock! (Let alone 100% allocation in the stock market)

fling93 writes:

Somehow I don't think 75% tax rates are very likely

I assume you're talking a 75% flat tax, not just a marginal tax rate?

I imagine that would most definitely push us to the right side of the Laffer Curve (and I don't think we've ever been close yet).

Boonton writes:

True...one thing the gov't could do is tax 401K plans that were in existence before a certain date (say 2010) at a higher tax rate. From a supply side point of view there is nothing the 401K holder would be able to do....his balance is his balance.....

I don't think the gov't will resort to such drastic measures but I do think it's premature to assume 401K's will be safe from taxation in the future. The arguments in favor of taxation will be that it is unfair for the entire burden of caring for retirees to fall upon the workers & 401K holders benefited from the gov't sponsored subsidies ever since they were created.

This doesn't mean that it won't be better to retire in 2035 having contributed to your 401K during your working life. It certainly will be. This does, IMO, mean the returns on 401K's will almost certainly be less than people are projecting when they try to compare them to SSI's 'returns'.

Now another argument is what type of returns can 401K's really generate for everyone. I think they would have to be no more than the overall economic growth rate.

Luke writes:

Boon, I didn't read the entire exchange between Eric and yourself, because I saw it drifting largely off-topic.

As far as the argument of whether 401k plans, IRAs, etc are paternlistic as well...

These private plans are optional, not mandatory. SS taxes are mandatory, the government is taking them from you with or without your permission, and giving it back to you if you ask for it. In this case, the government is not *influencing* or *coercing* people to save, it's *demanding* that they save, and save in the way that the government demands.

Paternalism means the government provides for citizens' needs without giving them rights or responsibilities.

In the case of SS, the government *DOES* give you the responsibility (to pay your SS money),but does NOT allow you your rights to use that money as you wish.

In the case of IRA, the government does not give you anything, it merely allows you your rights to use your money as you wish, and you keep the responsibility of providing for yourself in retirement.

Social Security is not paternalism, it's forced use-of-property, with a good dose of redistribution mixed in, but that's off the paternalism subject again.

IRA is not paternalism, it's freedom of use-of-property.

Your suggestion that IRAs and 401k's are a form of paternalism is rooted in the assumption that the money came from the government, and that it is gifting those account-holders thru tax breaks. The suggestion that SS is paternalism is only slightly more viable - it is based on the same assumption, but in the case of SS, the government takes the money by force (jail-time for un-paid taxes), and then gives it back.

Neither case is truly paternalism. But only SS violates an individual's rights to their property.

Boonton writes:
IRA is not paternalism, it's freedom of use-of-property.

Your suggestion that IRAs and 401k's are a form of paternalism is rooted in the assumption that the money came from the government, and that it is gifting those account-holders thru tax breaks. The suggestion that SS is paternalism is only slightly more viable - it is based on the same assumption, but in the case of SS, the government takes the money by force (jail-time for un-paid taxes), and then gives it back.

Luke, would you say farm subsidies are not a form of paternalism? After all, the gov't isn't forcing anyone to be a farmer or forcing farmers to grow a certain crop? Right?

The fact is that income is taxed at a particular rate. If person A's income is exempt from tax because he is doing something the gov't wants him to do, it means that everyone else is paying for that particular person's tax cut. Suppose the gov't said anyone named Eric would get a 5% tax cut. Certainly you aren't going to claim that such a policy is costless or that it's not paternalistic towards people named Eric because it's just letting them keep a bit more of their property!

Tax subsidies are still subsidies. Your property argument is a weak way to try to spin the huge amount of pork that is written into the tax code. While the benefits of the 401K are voluntary for those who want to participate, the costs are mandatory and fall on everyone who chooses (for good or bad reasons) not to participate.

Eric Krieg writes:

Boonton, if you can't see the difference between the government encouraging people to use 401(k)s through a subsidy and forcing people into the Socialist Insecurity system, then there really isn't much point in arguing further.

They both might be paternalistic, but they are on completely different orders of magnitude. It's the difference between a drizzle and a hurricane.

Luke writes:
The fact is that income is taxed at a particular rate.
And our entire income tax system is a complicated mess of pseudo-subsidies, yes. The definition of paternalism I'm using is one in which the government provides for needs WITHOUT giving citizens rights OR responsibilities. Nearly ALL of our current tax code gives citizens no rights - you comply or you go to jail, while at the same time giving us the full responsibility for the tax. It is not paternalism, but it's not freedom.
Tax subsidies are still subsidies.
But they are just that, subsidies, and not real paternalism. Farm subsidies are part of that same pseudo-paternlism as the rest of the tax-and-spend system, and so is Social Security.

In every subsidy from the government, the right of the citizens (tax-payers) to their own property (the production of their own labor) is violated to begin with, so that takes care of the 'no rights' requirement. But in every subsidized case, you have certain responsibilities to meet in order to receive that subsidy. You have to save your money in a certain account, or you have to take on a certain occupation, or you have to be above a certain age, or you have to be below a certain income level, whatever. None of these are real paternalism.

What we have is a complicated mess of trying to give people rights and freedom, while at the same time relieving them of their responsibilities. What has actually happened is the reverse - we've denied people their rights and imposed more responsibility onto them. 4 of my friends and I will be paying for Grandpa Joe's retirement income, and neither Joe nor the 4 of us have any say in the matter.

The income tax system is a huge mess that makes tax loopholes and subsidies rampant, and at the same time, obfuscated.

Joe B writes:

I'm not comfortable with Diamond's paternalistic assumption about government stepping in when people do not act optiamally for themselves.

This country would be much better off and have a much better allocation of resources if fewer of its citizens depended on the government to look out for their best interests.

Diamond's contention also assumes that the handful of bureaucrats in congress are better able to determine and pursue what is optimal for individuals. The most salient contradiction in this assumption is that the bureaucrats in congress don't seem to believe that what is optimal for their constituents, namely, social security, is optimal for their own retirements.

What is it about being a congressman that alters retirement optimization and makes it different from the retirement of the private citizens they represent? What is it about being in congress that suddenly enables them to best determine what is optimal for their own retirements, even though the day before they were voted in they were considered unfit to do so?

Boonton writes:
But they are just that, subsidies, and not real paternalism. Farm subsidies are part of that same pseudo-paternlism as the rest of the tax-and-spend system, and so is Social Security.
Boonton, if you can't see the difference between the government encouraging people to use 401(k)s through a subsidy and forcing people into the Socialist Insecurity system, then there really isn't much point in arguing further.

In both cases gov't is taxing everyone else at a higher rate in order to provide the benefits of SSI or 401K. To the degree that 401K's produce better 'returns' than SSI then the degree of paternalism is all the higher since everyone else is taxed just that much more to compensate for the lost revenue.

To bring this point home even more, imagine gov't abolished the income tax in favor of a national sales tax. The effect on Roth IRA's would be devastating because their main benefit (no taxation) is now gone. Retirees will be paying tax on their Roth IRA & 401K money as they make purchases! If the gov't wanted to make such a change in tax systems neutral, it would have to end up giving 401K holders actual payments to compensate them for their lost advantage. What better illustration of 401K's paternalism do you need? Who would pay for those payments? Why the taxpayers and it wouldn't be voluntary!

Eric Krieg writes:

From the government's perspective, they are no different if the costs are the same. But they are A LOT different from the individual's perspective. One allows freedom, the other does not.

BTW, I don't think that there are many 401(k) savers that would be against a national sales tax just because of their 401(k). A national sales tax would be very good to savers. It would be murder on all the people with maxed out debts.

Boonton writes:

If the goal of saving was saving in and of itself, then a national sales tax would not be an issue to 401K savers. On the other hand, if people saving for retirement actually expected to have a retirement then a national sales tax would be harmful. By definition it would reduce the amount of consumption a 401K retired person could enjoy.

What you are missing Eric is that the 401K system does not allow for more net freedom. It gives those that use 401K's more freedom at the expense of those who don't. This is like saying sugar subsides give farmers who want to grow sugar more freedom. I could also point out those receiving SSI checks enjoy the additional freedom gained by those monthly checks as well.

Eric Krieg writes:

>>I could also point out those receiving SSI checks enjoy the additional freedom gained by those monthly checks as well.

Yep, we Gen-Xers are going to take our negative return on SS to the bank!

The 401(k) is paternalistic in only one sense. 401(k)s can only be used for retirement, and perhaps a few other specific situations that I am not all that familiar with (first home?).

I would prefer a 401(k) like savings vehicle that could be used for ANY purpose. That would maximize freedom.

Of course, so would just making interest on savings tax free, which is as it should be.

Boonton writes:

SSI's returns require you to know how long you will live. Since you could get struck by lighting the second after you get your first check, there is no way to be sure you'll get positive or negative returns. What is clear, though, is that regardless of what your eventual return will be the first check you get will increase your possible consumption for that amount thereby providing you with an increase of freedom.

Again I ask why should income earned from savings be tax free when other types of income (such as by working) are taxed? The double taxation argument falls apart as soon as you give it even a little scrutiny. What you are left with are paternalistic arguments (mmm...savings are good...okayyyyy).

phwest writes:

Interesting discussion, but I wanted to make a couple of observations :

1) Assuming no change is tax policy and a flat tax, the 401(k) and Roth tax effects are identical - P(1-taxrate)*return (tax free) = P*return*(1-taxrate) at retirement. Thus the 401(k) subsidy is essentially the non-taxation of investment earning.

2) I am in a situation where I will probably start shifting to the Post-tax Roth-lite 401(k) option because of the current reality that taxes are not constant over time. Most common deductions are stage-of-life dependent. When I was single and living in an apartment my marginal rate was 28%. Now despite a gross income roughly double what I was making 10 years ago, my marginal rate is only 15% due to mortgage interest and dependent deductions. 15 years from now, I will have lost the bulk of that, and my marginal rate will be back up to 28%. And it is at the margins that the relative penalties kick in. If you're not careful, you can be putting money in a 401(k) at 55 and getting a 15% break, only to be taking it out at 28% 10 years later - It takes an 8% return to make up for against non-taxed prefered savings, much less a Roth.

3. For a very young investor, the 401(k) has one big advantage - you can gamble with the house's money. If I was 25 and single working for an Enron it would be entirely rational to overload on company stock in the 401(k) - if the company tanks I can make up for it later, and if the company is successful I can be a Microsoft millionaire before I'm 40.

Eric Krieg writes:

>>there is no way to be sure you'll get positive or negative returns.

You are obscuring the fact that the PROMISED benefits are less now than they were in the past, and more importantly, the taxes are much higher than they used to be. Also, the projected deficit is such that the promised benefits are never going to materialize.

So, on that basis, the return on SS is negative, to me. I am going to lose money on the deal, in all likelihood. Probalistically speaking.

Boonton writes:

Promised benefits, to my knowledge, were only cut once...in the early 80's and that cut amounted to simply holding off on some inflation adjustments. Hardly anything compared to Nixon's increase of benefits for not good reason at all beyond getting reelected.

Whether positive or negative, the fact is the person getting an SSI check has more freedom by definition. Likewise it's silly to say 401K's expand freedom when you are just looking at the side of who benefits.

John J. Olson writes:

Gentlemen, please let me throw some facts in with theory. Boonton, it is entirely possible to pay a higher rate of taxation on a tax-deferred account than you would have paid while accumulating that account, due to the required minimum distributions at ages above 70 1/2 and the progressive income tax structure. By the time you're in your eighties, the required minimum distributions get higher and higher. If you have a big 401(k) by then, these required minimum distributions can push you into the 35% income tax bracket even though you were in a lower income tax bracket at younger ages. At these levels of adjusted gross income, 85% of your Social Security benefit will be taxable even though half of the money from it was subject to income tax when the SSA deducted it from your paycheck. Hence, you are being taxed twice by the IRS on nearly half of your SS benefit.

Regarding savings, the American savings rate was 8-9% through the stagflation days of the 1970's and was 6-7% up to the mid-1980's. Today it's an anemic 2%. Why? Because the payroll tax, which is the largest tax for 70% of US households, doubled in the mid-eighties.

Congress can offer all the tax sweeteners it likes by raising the 401(k) contribution limit, by granting extra deductions for those over 50, by creating the Roth IRA. But, these will be weak incentives for people in low income tax brackets, which is most people. They're still paying more money in payroll tax than income tax. After decades of this, they certainly feel entitled to generous SS and Medicare benefits. The fact that their money went out as fast as it came in, a fact that Congress has continuously covered up with a fictitious trust fund, does not inhibit their demands.

The default answer and the most likely prospect is a big increase in the income tax, especially in the upper brackets. This is because first, the Treasury will have to redeem the bonds it has sold the SS Trust Fund through general taxation and second, Congress has no other source of money adequate to the purpose. How will Congress save Social Security? By taxing the hell out of your carefully-husbanded 401(k).

Eric Krieg writes:

>>How will Congress save Social Security? By taxing the hell out of your carefully-husbanded 401(k).

A sales tax won't do it, huh?

What scares me is not increased income taxes neccessarily. It is a wealth tax.

Boonton writes:

So when we attempt to measure the returns earned on 401K's, we should take into account the fact that 401K's are likely to end up being taxed at some point....this is just an indirect way of increasing income taxes since those with the largest 401K's will have had the largest incomes (income being both wage income and investment income).

This brings me back to my suggests for sensible reform:

1. Index the retirement age to expected lifespan going forward.

2. Removing the payroll tax on earnings would result in excessive marginal tax rates but what could be done is to losen the cap. Let the portion of income over $88K be taxed at 3-4% for SSI. Use 1 or 2% for increased benefits on those who had higher incomes and the balance to help fund the program...including reducing debt today in anticipate of the baby boomer crunch.

3. Find ways to encourage willing and able seniors to delay retirement & continue working. For productive seniors, this would be a double win because their income would be higher & gov't revenues would benefit from additional SSI & income taxes.

4. Increase immigration for younger people.

5. Rework the healthcare system into a universal voucher funded by a dedicated tax.

Social Security is already on the cusp of being fine. Projections of bankruptcy require decades of below average growth while good growth will make the program sound for over 75 years (who knows what the world will be like then? How different was 1900 from 1975).

John J. Olson writes:

Boonton, I don't see the connection between your assumptions and your claim for the actual health of Social Security so I cannot address them. However, your previous post had these cures for SS if indeed it's in actuarial danger: 1. Raise the retirement age (cut benefits), 2. Raise the taxable wage base (raise taxes), 3. Encourage old people to work (pay more taxes and get lower total benefits), 4. Encourage immigration of young foreigners, and 5. Nationalize medicine (raise taxes). Each proposal other than increased immigration is a tax increase.

Let me examine #2. You suggest raising the taxable wage base by taxing earned income above the current approx. $80K base, a tax of 3-4%, then use part of that to subsidize people with lower average indexed monthly earnings. That would be like the current system, only more so, since the current system replaces the first block of earnings at a higher rate than the second block above it, and does not replace any earnings at all above the taxable wage base. Hence, SS's tax is flat up to the wage base, but it is regressive (the lower the amount, the more of it is replaced) in its benefit formula.

You're right that the only alternatives are to cut benefits, raise taxes, or some combination of the two. But, you are proposing to turn a purported insurance plan into an income redistribution plan. Is this fair? Moreover, won't it aggravate the moral hazard of too many people relying too heavily on Social Security instead of accumulating more personal savings?

Further, Boonton, proposal #2 conflicts with proposal #3. Today, most SS recipients take a partial benefit at 62 since it takes 13 years for the higher benefit at 65 to make up for those three years of early benefits. Consequently, their benefits are reduced by $1 for every $3 they earn above a low level about $12,000. If you would tax a 62-year-old doctor more heavily on his earned income, you're giving him an incentive to retire that much sooner.

Boonton, if people are to be coerced into providing for retirement, wouldn't it be more equitable to coerce the average taxpayer into funding his own retirement than to coerce the above-average taxpayer to fund the retirement of people who refuse to provide for themselves?

Only about half of the people between 25-64 have 401(k)'s or any retirement savings plan. Only 10% of 401(k) participants contribute the maximum. Most taxpayers qualify for deductible IRA's but fewer than 2% of tax returns include any adjustment to income for IRA contributions. Saving for retirement is something most people don't take seriously. If we're going to put a gun to a man's head (which is how laws are ultimately enforced) to make him pay for somebody's retirement, wouldn't it be more fair to make him pay for his own than for somebody else's?

Boonton writes:

John,

The passage about SSI's actual health is a clickable link. I'll let that article stand or fall on its merits. I don't see any obvious flaw in either its numbers or its reasoning. The most potent part is the low growth assumptions used to depict SSI as on the verge of bankruptcy. If those assumptions are true then the next 20, 30 40 years onwards will not be that pretty for the stock market either....so much for 401K's being the magic bullet to save retirement!

For some other points:

1. Obviously some older Americans cannot, should not be working while others are able to work. As lifespans increase there will be increasing portions of the older population who are perfectly healthy at 75....while there will be some 60 year olds who are more like 90 year olds. What's interesting about the European System (see http://econlog.econlib.org/archives/000451.html) is that they have set things up so that the least productive seniors are encouraged to get out of the labor market while more productive ones can stay.

2. If retirement age is indexed to increases in expected lifespan, how are benefits being really cut? Considering that SSI originally gave you something like -1.5 years of expected benefits how bad is the return really?

3. Taxing higher incomes to fund SSI does create some incentive problems. That's why I wouldn't do it at the full 14% tax that people under $80K pay. I didn't say there would be no balancing issues here.

4. SSI functions as an income redistrbution program. That works nicely for its primary purpose; that is to protect the eldery from grinding poverty. I'm not particularly intersted in making sure Bill Gates allocates enough to protect himself against a 'huge downfall' in his retirement living standards (say to just a $1M per year). So while it may be politically easier to force people to save only for themselves, IMO it doesn't address everything.

John J. Olson writes:

Agreed, benefits are not being cut, merely held stable, if indexed to increasing longevity. When you say to a SS taxpayer that he or she will get benefits at 70 instead of 67 because he or she is likely to live to 83 instead of 80, that doesn't look like a cut to anybody except the taxpayer.

You have said that the European system sets things up so that the more productive senior citizens can stay in the work force while the less productive ones can start drawing retirement benefits. But, why should that more productive senior citizen keep working? He has a strong financial incentive to take his retirement benefits as early as he can. In most European countries, you don't get higher benefits by waiting longer than ASAP to take your retirement benefits, while the payroll taxes on your earned income are high to pay for everybody else's retirement. When you add the payroll tax to the foregone income of missing that first year of retirement benefits, you lose an amount of money equal to about 75%-82% of that year's earnings in the UK, France, Italy and Belgium. In the Netherlands, it's 141% since your retirement benefits are free of income tax. (Source for these figures is "Gray Dawn", by Peter Peterson, a former U.S. Secretary of Commerce)

Under these incentives you'd be crazy not to retire as early as you can, whether you're among your country's most productive middle-aged or not. So, that is what they do. The Western Europeans are rightly proud of their average longevity as a reflection of their fine health care systems and this longevity has increased markedly in the past few decades. Yet, in spite of that admirable increase in elder health, their workers retire much younger than they used to.

Again from Peterson: Between 1960 and 1995, average retirement age for German men fell from 65.2 to 60.5. In France, from 64.5 to 59.2. In Italy, from 64.5 to 60.6. In the UK, from 66.2 to 62.7. (In the USA, from 66.5 to 63.6.)

If those 60-year-old Europeans are getting healthier, then why did the share of men 60-64 in the work force drop from 70% to 32% in Germany in the last 25 years? In France, it went from 67% to 22%. Because it isn't a matter of health, it's a matter of incentive.

Peterson did not examine whether the more productive middle-age workers were more or less likely to retire ASAP, but under a progressive income tax structure the higher-paid face a bigger tax bite than the lower paid. So, why should they keep working?

You have said that it is "politically easier to force people to save only for themselves", but I didn't say it was politically easier. My question was whether it was more fair and just that the first claim on a worker's paycheck belongs to himself or to somebody else. Would you address that question? Since you can't eat your cake and have it, too, you seem to be saying that those who have already eaten their cake are entitled to a slice of somebody else's.

Boonton writes:
My question was whether it was more fair and just that the first claim on a worker's paycheck belongs to himself or to somebody else. Would you address that question? Since you can't eat your cake and have it, too, you seem to be saying that those who have already eaten their cake are entitled to a slice of somebody else's.

Clearly the worker should have first claim to his own paycheck. The question you are really asking is 'why is paternalism ever acceptable'. It's not really giving a person anymore freedom to tell them they must set aside 10% of their money until they are 65. What if they die at 64? What if they plan to kill themselves at 65 after living their life to the max? Any good libertarian worth his salt should immediately be able to shoot down forced savings as easily as he would shoot down SSI, welfare, food stamps, etc.

In my previous posts I've explained why I believe paternalism is sometimes acceptable even though market solutions are usually perferable. Experience has shown that many will neglect to save properly for retirement (and many at the bottom of the ladder simply cannot). The 'market lessons' are exceptionally harsh & provide little opportunity for the victim to correct his mistake. As a society we have decided to make our own solution rather than accept what the market says.

John J. Olson writes:

Boonton, what you call paternalism is enforced by coercion whether the first claim on a worker's paycheck is paid to himself or somebody else. You find this coercion, which is ultimately enforced at gunpoint, preferable to letting improvident people suffer the consequences of improvidence.

But, this creates a moral hazard. Why should a man who might provide for himself make any sacrifices to do so when others will do it for him or coerce third parties to do so? How can you encourage thrift and self-reliance by fining those who practice them, to grant bonuses to those who do not?

Just about every American worker has some type of retirement savings plan available to him. Of those covered by 401(k) plans, only about half participate. Most taxpayers are eligible to contribute to deductible IRA's and nearly all are eligible for Roth IRA's.

To fully fund a $250/mo. IRA, the average worker would have to give up cable television ($60), his cell phone ($40), brown-bag lunches instead of eating at McDonald's ($70), and drive a $20,000 car instead of a $25,000 car. Why should he give up all these things he can enjoy today to have a prosperous retirement? Will taxing him more heavily to provide for others teach him to save more to provide for himself? Or will it teach him to try to get the government to redistribute income to him?

The 2004 Retirement Confidence Survey by the Employee Benefits Research Institute says that 68% of Americans are very or somewhat confident of being financially comfortable in retirement, even though 45% of them have only $25,000 or less in total financial assets. More than half expect to work until or beyond age 65 even though most Americans today retire at 62. Two out of five recent retirees say they were forced to retire earlier than planned due to health problems or layoffs.

Surveyed workers gave a variety of reasons for not saving more. About one-third expect to get money from an employer. Others are relying on God, an inheritance, saving later, and Social Security. If you want to know what God thinks of spendthrifts, you can find it in Proverbs, but if you want to know what other people think of the respondents' plans to retire on Other People's Money, ask the other people.

Boonton writes:
Boonton, what you call paternalism is enforced by coercion whether the first claim on a worker's paycheck is paid to himself or somebody else. You find this coercion, which is ultimately enforced at gunpoint, preferable to letting improvident people suffer the consequences of improvidence.

This is the libertarian argument against paternalism....if someone didn't save then let him starve to death unless his family, friends or charity are willing to give him food. This is a good argument (I don't have any objection to people who slack off on their job getting fired, people who max out their credit cards having to declare bankruptcy etc.) In some circumstances, however, we have decided to reject it.

If you want to argue against paternalism then you should be consistent. 401K's are just another type of paternalism. They appear to be less of a stick & harm no one but the same could be said of protectionism.

The people harmed by 401K's don't see it because they get no bill that says 'your taxes are $X higher because we are granting favorable treatment to those who use 401K's,...you have to pay AT THE POINT OF A GUN(tm)'.

So what is to be gained by 'forced savings'? SSI is a society wide risk pooling scheme (as is all insurance). Your get something as long as you are alive. 401K's are quite variable. You could retire quite wealthy or quite poor depending upon how you managed your 401K. In either case you aren't guaranteed any income unless you choose to buy an annuity (which are currently complicated and filled with excessive fees, hidden costs & so on).

SSI isn't supposed to provide for retirement but simply provide a base regardless of your circumstances. Whether or not a person wants to provide for a retirement above that base should be their call. Imagine a policy is implemnented requiring 4% of your income to be saved. Should we really care if Bill Gates is setting aside $400M a year so he can provide himself with a retirement income of $40M a month? If he blows all his money & ends up living on just $75K per year it will be quite a drop for him but not anything that should concern society as a whole. If he was living on $750 a year, however, that would be a different story because we have decided that we do not want to accept people living on the streets, begging, starving to death etc.

(Some may object that the US does have homeless people and beggers...at least in its major cities. That's true to a degree but regardless the US has choosen not to accept the level of that sort of thing that other countries...like India...have).

Eric Krieg writes:

B, regarding that future tax rate, yesterdays WSJ has a book review of "The Coming Generational Storm". In it, the authors project a future tax rate of 70% if no changes are made in the current SS system.

Coincidence, huh? I know, we'd have to look at all the assumptions and whatnot, and I don't believe that there is a snowballs chance in hell that the current system will survive intact. No way is there going to be 70% tax rates anytime, ever. But its funny that they predict the exact tax rate that you say will cause 401(k) savers to lose money.

Boonton writes:

I'm curious to know if the 70% belongs to SSI or Medicare also? Also when does the 70% hit? 2020, which is relatively close enough to extropolate from the current trends or 2050 which is a half century away?

Eric Krieg writes:

B, you are going to have to buy the book!

I am sure it was all generational transfers. I believe that the extrapolation was "only" 75 years, so say 2080 or so. I agree, such a projection is pretty preposterous, but it is a starting point for argumentation.

What was interesting to me was that the "crisis" doesn't go away when that baby boomers die. I was under the impression that they were "the pig in the python" and that, once they were all dead (the sooner the better ;)) the problem goes away. But in 2080 the projections are still for more retirees than people under 20 years old. That isn't because of the baby boomers, that's because of TODAY's 20 year olds! Or kids even younger.

Boonton writes:

So our retirement system, coupled with our improved healthcare has created a situation where there are fewer workers supporting more non-workers.

I suggest you take a peek at Doug Henwoods article at http://www.leftbusinessobserver.com/AntisocInsec.html

You'll notice that we are actually at a low point of 'non-workers' to workers. There's two variables at work here. Fewer workers means the non-workers must make do with a less ample retirement, workers must make do with reduced consumption or both. This will happen in a 401K style world just as much as it will in a SSI only world.

It is only logical, the goods and services consumed by non-workers must be produced by workers. In any given year there is a finite amount of such goods produced so anything consumed by a non-worker (child, retiree, unemployed person) must decrease the goods available for the worker to consume. There is no way around that logic.

As the US population ages, pressure will be brought down on retirements as well as workers. In an SSI world this would result in pressure to raise taxes and cut benefits. In a 401K world this would manifest itself in lower returns on 401K's for retired people and workers having to contribute more of their income to their 401K to achieve their retirement goals.

The extreme cases will not come to pass because society will find a balance between the two extremes. Indexing SSI to lifespan, for example, is one such act of moderation.

The other escape hatch is productivity. If the next 75 years turn out to be filled with fantastic increases in technology...it's possible that worker productivity will increase so dramatically that you can have the luxury of more non-workers to workers. If the average workers' income jumped from $35K to $95K, for example, he can easily support more SSI retirees without a tax increase.

Boonton writes:

Eric,

I skimmed thru the book at Borders (cost, $1.80 for mug of coffee!). One feature I found really intersting was that they looked at 401K's from the perspective of taxation of SSI benefits. It didn't occur to me that SSI benefit reductions should be included when one estimates their return on a 401K!

I wonder what your thoughts are on his call for a revised SSI with private accounts invested in a universal 'global stock/bond fund'? Such a reform would be ok by me, but I wonder how Wall Street will feel with SSI money not going into the endless number of fee burning mutual funds they create?

Eric Krieg writes:

Boonton, I am for any privatization that gives people personal accounts and invests in the private sector. The 401(k) does not have to be the model that privatization is built upon, although that is my preference.

Creating a global stock/ bond index fund is a pretty good idea. It minimizes costs, it minimizes the ability of the government to meddle in private markets, and it gets American money invested abroad, where investment returns might be a little higher over the long term.

Boonton writes:

The problem, IMO, with the 401K model boils down to a few things:

1. It is tilted towards those with higher incomes. Why should taxpayers pay higher taxes to encourage a $100K a year doctor to correctly plan for his retirement? If he screws up, it is unlikely that he will be living on the streets..so to speak.

2. It is a subsidy to mutual fund managers, financial 'advisors' and a host of other people who are basically parasites. These people typically charge fees that can end up eating 30% or more of a person's returns (many of the fees are hidden and almost impossible to discover). In exchange, these people end up producing returns that are actually below the market! (for a given level of risk, it's possible to appear to 'beat the market' if you're generating higher returns by investing in securities that have above average risk).

3. The authors have a dead on analogy; it's like the gov't is providing for retirement by driving people up to a casino, handing them some chips, and telling them to do their best.

If people want to play the market, they should do it with their own money. Tax payer money should only be invested in the 'global index fund' that the authors lay out. Basically this would simply be an index fund of all the world's stocks and bonds weighted according to their market capitalization.

For example, if Microsoft represents 1.2% of the total market cap of the entire global market then it would represent 1.2% of an ideal index fund. The fund would require administrators to do the number crunching and periodic rebalancing as different securities change in price. Also, there would have to be some descision making on the margins...such as do you include the shares of Afghanistan's stock market or wait and see if they really have their act together?

But such a fund, even though it would be gov't controlled & sponsored, would be insulated from politics. By law, no politician could influence the fund to, say, buy or sell RJR Tobacco or Playboy...the allocation would be taken from the market & investors would basically be 'betting on the world economy'.

Let's be clear, though, that's all anyone can bet on. SSI, funded thru taxation, is a bet that the economy of 2030 will be able to provide for the retired people in 2030. If a comet hits earth in 2029 & wipes out civilization it doesn't really matter if your savings was in SSI, a 401K or a modified SSI. Likewise, if economic growth is not strong for the next few decades the retirement of the boomers and post-boomers will be difficult regardless of what system is in place.

Comments for this entry have been closed
Return to top