Monday, April 7, 2008

Social security reform is possible

We have all heard the mantra that social security is going to hell. Whether this is true or not, it is now generally accepted that a fully funded social security system is superior to a pay-as-you-go system. The issue is the transition, as switching from the latter to the former leads to a generation missing out (the old one) or one generation force to save doubly (the young one). This is especially bothersome because of current demographic change.

The problem essentially is that while you have a Pareto improvement when comparing the two systems in steady state, such Pareto improvement is impossible during a transition. For example, the current old generation, whose retirement is currently funded by the young generation, would lose in a switch when the young ones suddenly contribute to their own retirement. Such a transition has somebody losing compared to the current status quo, so no Pareto improvement may seem possible.

Not so, say Juan Conesa and Carlos Garriga. If you combine social security reform with optimal fiscal policy one can obtain a Pareto improving transition. The key here is the ability to differentiate tax rates by age. Essentially, you want to subsidize labor income for older people during the initial phase of the reform, and then gradually get to a steady state schedule that assume almost identical tax rates across age groups. It is just the opposite for capital income tax rates. All flat and positive initially, and then decreasing with age in steady state, with subsidy for the older ones.

PS: This evening is the final of the men's NCAA basketball tournament. See what view you have for $275 tickets.


Anonymous said...

Pareto approves (of the policy, not the NCAA tickets). In any case, redistributive taxation is there to even out the luck people have, why not also take into account the luck of being born in a particular generation?

Bruce Webb said...

"Whether this is true or not, it is now generally accepted that a fully funded social security system is superior to a pay-as-you-go system."

Generally accepted by whom? Certainly not by people who have actually examined the numbers in context and are not blinkered by Chicago style orthodoxy.

Social Security 'crisis' is rarely defined in context. Under current projections by the Trustees Trust Fund Depletion in 2041 will require a reset of Social Security benefits to 78% of then current levels, themselves set to be 160% of those of similarly situated retirees today. Which is to say the check would still be 25% better in real terms than my Mom gets today. Should we decide today that this result is unacceptable we could backfill it with an immediate 1.7% increase in payroll. Or contrawise come to the reasoned conclusion that it is possible to grow the economy at higher rates than the ultimate productivity and Real GDP projections under Intermediate Cost of 1.7% and 2.2% respectively. Or of course some combination of better growth and slightly higher taxes.

From a societal perspective there is financially no difference between fully funded and paygo. In each case you have a claim on future productivity on behalf of future retirees. The preference that that claim be exercised through dividend payouts and capital gains rather than by a governmental check is just a preference based on ideological hostility to the New Deal.

No one has successfully met the No Economist Left Behind challenge. You simply cannot make a system of private accounts (your 'fully funded') work under Intermediate Cost economic assumptions, you can only slice the gains of an economy growing at ultimate 2.2% Real GDP so far.

Frankly there is a lot of bait and switch going on with 'crisis' numbers and dates drawn from one set of books and private accounts proposals implicitly using alternate and higher growth models to achieve their 6.5% real assumed returns. Which is why these discussions never seem to get around to discussing actual assumed levels of future growth. The Trustees have provided us with a model that shows Social Security fully funded and in the 2008 Report over funded through the 75 year actuarial window. It is called Low Cost. This model may be too optimistic in some places (fertility) but that is largely offset by places where it is too pessimistic (immigration). In any event it establishes a budget for transition plans to 'fully funded'. Can you actually provide a better result for future retirees at less cost than the 1.7% payroll tax fix needed under Intermediate Cost or without projecting ultimate productivity numbers at or above Low Cost, i.e. 2.0%? If not you are not making an economic argument but simply exercising a specific ideological preference. And more and more people are waking up to the fact that 'crisis' has been sold by scare tactics and the proposed 'reforms' actually do nothing to improve the bottom line of retirement security.

Economic Logician said...

Bruce, just think about a simple textbook model of overlapping generations. In a PAYG system, the old generation is fed by the young. In a fully funded system, the old live from savings plus return. For people to prefer the PAYG system, you need to have a population growth rate (from immigration and fertility) higher than the return on savings. This will not happen in the foreseeable future for most industrialized countries.

In addition, you have various incentives playing against PAYG: labor income taxation reduces labor supply, and the lack of an incentive to save lowers capital accumulation. All this leads to lower output (and income) compared to a fully funded system.

So, given a population growth, you can choose which system to adopt. But once you have one, it is difficult to change to the other without hurting a generation during the transition. The Conesa and Garriga paper shows it is actually feasible to obtain a Pareto improving transition from PAYG to fully funded in the current US context.

Bruce Webb said...

That seems to be kind of a cartoon version of what 'savings' really is. In the real world you give money to the bank and they spend it. In return they promise to pay it back with some rate of interest. But really they are simply making a bet on future productivity, if the economy doesn't grow at rate fast enough to justify that interest rate the bank loses money. Now in the case of a FDIC insured bank the depositor has every reason to believe there is 'money in the bank' but this is really a fiction. The actual cash dollars that will be used to fund those retirements will have to be extracted from then current economic product.

People like to think of banks and other financial institutions as if they were the actual counterparts of Gringott's Bank in the Harry Potter series, endless levels of individual vaults with your assets locked inside. Just as they like to think of the dollar bills in their wallets as being 'real money'. In reality everything is just a claim on future productivity, those dollars are mixed together and put to work. In the case of a US dollar bill that isn't much of a claim on a week to week basis, in Zimbabwe things are a little different. But ultimately it is all a bet on the future economy.

To paraphrase Reagan: 'a simple textbook model is not the solution, a simple textbook model is the problem'

"For people to prefer the PAYG system, you need to have a population growth rate (from immigration and fertility) higher than the return on savings. This will not happen in the foreseeable future for most industrialized countries."
You have tried to disconnect two things that are bound tightly together. First of all you are totally neglecting the vital issue of increasing productivity per worker. I see people claiming that the gains from automation are largely in the past, frankly those people need to get out more. Second you are skipping over the fact that you cannot ultimately get a better return on savings than the overall economy produces. You can't just reduce the whole question down to a single question of covered worker ratio. That is good enough for the rubes but doesn't work for even a layman who was spent any time thinking about what money and investment are and are not.

"labor income taxation reduces labor supply, and the lack of an incentive to save lowers capital accumulation. All this leads to lower output "
In theory, in simple textbook theory, but you are building in some assumptions about labor compensation and marginal productivity that not everyone finds persuasive. Once again I find your argument to be in the final analysis ideological, a preference for markets and a solution in search for a justification.

I came up with a ditty a decade ago:
'If private accounts are possible they won't be necessary.
If private accounts are necessary they won't be possible'
I see no reason to revise that at all.

Privatizers' theory works okay given their assumptions. It is just that they can't actual deliver results for lower quintiles of workers, when put to paper the numbers just don't run. And certainly not under SS Intermediate Cost assumptions.

There is an honest debate to be had for implementing some sort of income related phase out plan which would allow workers earning over the median to reallocate some of their deductions to other asset classes, perhaps through some vehicle like the federal Thrift Savings Plan. But since such a structure doesn't allow the Cato types to kill off Social Security for all time, a goal of some since implementation in 1936, I doubt we will see any such debate any time soon. Instead we will see a series of theory driven, largely number free arguments with the goal of rolling back the New Deal. To quote from one of my early blog posts on this (I maintain an intermittent Social Security blog) "This site is all about the numbers. You get links to every Social Security Report from 1942 to 2005, you get breakouts to particular tables from all Reports from 1997 to 2005, you get some explications of what those tables mean. The numbers are important, if you are going to participate in the debate over the future of Social Security you need to understand them, you need to understand their implications, you need to be able to measure them against the numbers you read in the paper every day. Because oddly enough this debate is not about numbers and in most respects it never has been."
True as it ever was.

Economic Logician said...

Bruce: did I ever mention money or banks in my post and comments? No. Because it is not a problem of money and banking. It is a problem of distribution of real resources across time and across concurrently living generations. And because this is not an endowment economy but a production economy, how it is distributed matters for the size of the pie and when it is available.

In my analysis, productivity increases at the same rate in both scenarios, thus it is immaterial in determining which is best. But it matters for the transitions from one to the other, and the referenced article takes that into account.

Anonymous said...

bruce webb: try to think in terms of goods, not money. In a pay-as-you-go system, those working get some of their goods taken away for the benefit of the retirees. Why would the former want to work hard? And would they save anything for later? There will be new workers when they retire.

In a fully funded system, they would get to keep all their goods, and they would make sure to save some of them for later. That is how you get more capital in the economy. Thus more effort in work, more capital: more goods, and it is easier to get enough for retirement.

Bruce Webb said...

Vilfredo you are still talking in cartoon terms. The world doesn't work in a way that you just pile up Galleons, Sickles and Knuts in your own vault at Gringott's and so "keep all their goods", ultimately the mayo goes bad and even the pickles look a little suspect.

You are appealing to a little piece of parlor psychology that has little roots in actual economic behavior. If in 2024 I am sitting at home or in a bar and drinking a beer or eating a sandwich and paying for it with either a Social Security check or a withdrawal from my IRA I am simply making a claim on current productivity based on past productivity. Indeed some people would have made life choices that laid to a larger potential yield from that IRA than a return from paying FICA but in the end it is all a transfer from future productivity to former producers. The rest is pretty much just bookkeeping.

Steve Selengut said...

A Capitalist's Social Security, 401(k), and Retirement Plan Reform Program

What if there was an easy way to implement a whole new approach to retirement funding, pension planning, and Social Security? Would the politicians be interested? Let's find out.

What if the new plan actually reduced payroll taxes, cut prices, created jobs, increased salaries, raised shareholder dividends, partially funded decreased healthcare costs, and was available to everyone?

Sound too good to be true, but it's actually doable. The reasons for the present system's failure are mostly political; the solutions are clear, practical, and non-partisan. What we want is a less expensive system for assuring that everyone is able to retire with an adequate income, higher than that provided now by Social Security.

What we need is a simple program, part mandatory and part voluntary, using experienced trustees who operate within the strictures of the prudent-man rule--- a risk-minimizing legal doctrine that restricts investments to those that seek reasonable income and preservation of invested capital--- SIBORAP Tier One investments.

The 2007-2008 stock market correction and credit crisis laid bare the weaknesses of all self-directed retirement accounts. First of all, they are not (and never were) pension plan equivalents. They were cheap-to-provide replacements for fully funded defined benefit pension plans--- supplemental programs at best.

Next, inexperienced investors were provided with an array of far-too-speculative investment options, and little if any training in basic QDI (Quality, Diversification, and Income) investment principles. The mutual fund industry was allowed to monopolize the self-directed plan market place.

Third, most participants thought of their programs (401(k)s, IRAs, ROTHs, SEPs, SIMPLEs, etc.) in guaranteed pension plan terms. They were encouraged to do so purposely by mutual fund distributors and inadvertently by uninvestment-educated employee benefit representatives.

If good news ever becomes an actual news story again, people would realize that both defined benefit pension plan and guaranteed fixed annuity contract payments were maintained throughout, and in spite of, this terrible financial environment. Why not deal with Social Security in the same manner?

A Social Security Retirement Income Annuity, or SSRIA, invested 70% or more in government guaranteed securities, could be phased in quickly as a mandatory replacement for the existing Social Security program. The personally owned SSRIA would also become a voluntary investment option for all self-directed programs and a guaranteed safe savings vehicle for after tax discretionary dollars.

These are the bare bones parameters of the new program:

SSRIA contracts will be provided by newly formed subsidiaries of established insurance companies. They are deferred, fixed-income-only annuities with no commissions or fees paid by participants or employers. All companies would provide identical products, insurances, and maturity options. A minimum of 150,000 new jobs could be created.

The contracts would include $10,000 of term life insurance, provide for retirement at age 60 or above with just two immediate annuity options: life and joint life. No variable account features, or withdrawals, would ever be allowed, and all SSRIA retirement payments would be absolutely income-tax-exempt at every political level.

SSRIA providors would receive an investment management fee of .85% of the Working Capital under management, emphasizing the importance of both income generation and preservation of capital. Participant account statements would reflect ever-increasing cash balances, growing at annually adjusted, contractually guaranteed rates

Providor operating profits would be distributed 70% to parent company shareholders and 30% to fund a trust for retiree health care benefits. An associated tort reform bill would cap jury awards and attorney fees for personal injury lawsuits against all health care providors.

SSRIA mandated contributions would be capped at 3% of pre tax total employment compensation; an additional 2% of pre tax earnings could be contributed voluntarily. Voluntary contributions to an employee's SSRIA would be a required investment option of all self-directed employee benefit programs.

There would be no employer contribution to individual SSRIAs. Employers would be required to use their savings in any combination of these options: increase non-executive salaries, hire additional workers, reduce consumer prices, and increase shareholder dividends.

Employees earning total compensation in excess of $1,000,000 would pay 10% of the excess directly to the retiree health care trust. All special compensation arrangements, including stock option plans would be banned. Bonus payments in excess of 20% of base pay would be pooled, and divided among all employees and shareholders, dollar for dollar.

Employees would be assigned randomly to qualifying SSRIA providors, one contract per person. Self-employed persons, dependent spouses and children, would be eligible for SSRIAs, and would be assigned to a providor by the Social Security Administration.

The Social Security Administration would oversee the operations, pricing, and investment practices of SSRIA providors, qualify companies wanting to become providors, and implement the transition from the existing program to the new. The process could take up to five years, unless peace breaks out in the Middle East.

The transition to the SSRIA program would commence immediately, starting with employees under age thirty. Existing Social Security accounts would be frozen. Balances would be applied 50% in cash as an SSRIA deposit, 20% to the retiree health care fund, and 30% as a Federal Income Tax credit. Older employees would have proportionately larger direct credits to their start up SSRIA accounts.

One other thought: All active government employees at all levels, elected, appointed, or hired, would be transitioned into the new SSRIA system.

OK, there it is, a viable first step change plan that most of us would go for. Call your representatives, newspapers, and favorite radio talk shows. Hey, it's our money; let's keep it that way.

Steve Selengut
Professional Investment Management from 1979
Author of: "The Brainwashing of the American Investor: The Book that Wall Street Does Not Want YOU to Read", and "A Millionaire's Secret Investment Strategy"