DC and its think tanks are selling the idea that Immigration Reform will help fix Social Security.  For example, "Improving Lives, Strengthening Finances: The Benefits of Immigration Reform to Social Security".  Without making any statement on the merits of Immigration Reform, there seems to be little actual research that supports the idea that Immigration Reform will improve the outlook of Social Security. 

There are not any reforms on the table for the Social Security Administration to score.  But I make the statement based on information that the SSA has already provided.

First, undocumented workers provide a stable source of unattached revenue.  When Americans pay payroll taxes, the revenue creates future costs in the form of promised benefits.  Undocumented workers cannot collect Social Security benefits.  Millions of undocumented workers pay payroll taxes by either using someone else’s Social Security card or by using a false name and Social Security number.  The Social Security Administration estimated that this revenue exceeded 10 billion dollars in 2010.  Losing free revenue cannot help Social Security.

Second, the demographic mix of workers affects Social Security.  If the immigration reform brings in a higher percentage of single workers, or high-paid workers, or workers that work longer than 35 years then immigration reform may help improve the picture of Social Security.  None of the media reports suggest that the system will enjoy more profitable demographics.  And to be clear, improve the picture of Social Security and fix Social Security are trillions of dollars apart. 

Third, immigrants tend to be a very poor demographic to add to the mix of Social Security because they do not have parents that depend upon the system.  Social Security is an electorial priority.  Voters elect Congress which sets the payroll taxes, which dictates the level of benefits.  We are adding an audience that has no value to parental support.  This change will overtime work its way into our representative government.  Unless the individual prospects of Social Security are dramatically improved for workers - we should expect immigration reform to negatively affect support for the system.

What supporters of Immigration Reform are saying is that Social Security will enjoy a boost in the number of workers.  More workers equals more revenue - but they ignore the cost of taking that revenue.  It may help Social Security in the short-term, but longer term it will make the system even more unworkable.  This is not a good trade-off for the system or the people who will depend upon the system in the future.

The point here isn't the immigration reform is wrong, but the people who tell you that Immigration Reform will fix Social Security are wrong.

The debate about Social Security reform isn’t serious.  It is noise filled with statistics that paint a picture of convenience.  This piece comes from testimony in a hearing about Social Security reform before the Ways And Means Sub-Committee.

Despite Social Security’s great success, its growth in lifetime benefits over time has been decreasingly targeted at its major goals. Even while programs for children and working families are being cut, combined lifetime benefits for couples turning 65 rise by an average of about $20,000 every year, so that couples in their mid-40s today are scheduled to get about $1.4 million in lifetime benefits, of which $700,000 is in Social Security.

What Are The Major Goals Of Social Security?

Social Security is suppose to be old-age insurance, and has virtually nothing to do with other programs for the children and working families.  What is really meant here is that the system does not serve the major goals of the person testifying.

What About The Growth Of Benefits?

In all honesty, this couple does not expect to collect $700,000 or anything close to it.

1. When he says ‘scheduled’ he ignores the fact that this couple expects to retire after the Trust Fund is exhausted which means that they will be subjected to benefit cuts. 
2. These people would likely trigger means-testing that would claw back benefits at more than 10% per year  
3. These people may die before collecting a penny

But the number sounds impressive.  Well, until you consider how much this couple has contributed.  In all honesty, the couple depicted probably deserves more than $700,000.  Someone born in 1970, making 2/3rds of median income (roughly 34K in 2010) will lose about $600,000 in savings to Social Security.  In case of this couple, two people were contributing, and both made more than median income.

The New York Times Room for Debate section covers an open discussion about increasing the amount of wages subject to payroll taxes, also called the wage base. 

The best material comes from Andrew Biggs, who offers a summary of the reasons not to raise the wage base.

1. Roosevelt wouldn’t want it, since the cap helps distinguish Social Security from a “welfare” program. Under his administration’s original plans, high earners wouldn’t even have participated in Social Security, much less paid multiples more in taxes than they’ll collect in benefits. Come on, lefties, let’s honor FDR’s legacy.
2. The cap isn’t unusually low: today, 85% of wages are hit by Social Security taxes. The average since 1935? Eighty-four percent.
3. While coverage has fallen from 90% since the mid-1980s, research points to health care as a major culprit. When employers’ health outlays rise they reduce wages for employees, with middle class workers hit the most. The solution isn’t to tax the rich more, it’s to address health care costs.
4. Other countries have lower caps: in Canada payroll taxes are capped at the average wage, in the UK at 1.15 times, and in Japan at 1.5 times the average. Social Security’s tax ceiling is 2.9 times the average wage in US.
5. It’s a fat tax increase. Today’s top federal tax rate on earned income is about 45%. Adding state income taxes boosts it to around 50%. Eliminating the tax max effectively raises the top tax rate by around 12 percentage points. If I wanted that I’d move to Scandinavia. Except their taxes are lower…
6. And that’s before we’ve done anything to fix Medicare and Medicaid. Think the Left won’t ask for higher taxes there, too?
7. And we wouldn’t even collect all that money. Economists Emmanuel Saez and Jeffrey Liebman — no conservatives — concluded that, due to income shifting and behavioral responses, we’d collect less than 60% of what static projections claim.
8. Using more aggressive but still reasonable assumptions regarding economic behavior, such as from Harvard’s Marty Feldstein, net revenue gains would essentially be zero.
9. And to the degree we did raise extra money, Congress would spend it — it’s not going to be there to pay the bills in the future.

One of the more dangerous arguments in the debate about Social Security is the growing belief that Social Security is part of the social safety-net that protects the poor.  This argument has bled into the lexicon of both parties, and become a standard for media analysis of any proposal to reform the system. 

There is one problem: Social Security isn’t a safety-net, nor is it designed to act as one. Yes, the formula rewards people progressively less as they earn more. Yes, the formula reduces benefits for people who saved for their own retirement. But nowhere does Social Security pay people based on need.

The formula has more than 2,000 rules which change the benefits based on whether you have kids or how many times you marry. The benefit formula rewards people for living longer. The benefit formula rewards people who work longer.  No where in these formulas does the system pay more money to people because they are in need.

The formula allocates the largest amount of resources to people who contributed the most in the past, live the longest, have the most qualifying ex-wives, and have the most children after the age of 65.  Social Security pays the most to someone like Pete Stark.  Who is Pete Stark? An ex-Congressman who is wealthy by Congressional standards.

He will collect the maximum payment allowed by Social Security.  He is apt to live longer than most Americans.  He married three times, giving the system three potential wives to collect survivor benefits from Social Security.  The last wife produced 3 children all of whom have been eligible for Social Security almost since birth. 

On the other side of Pete Stark, Social Security will allocate zero resources to people who worked for Central Falls, a small town in Rhode Island which faces bankruptcy.  Like many towns and municipalities, Central Falls did not put its employees into Social Security.  With their pensions gone in bankruptcy, the people who retired from the city will have significant needs and no way to collect from Social Security.  The reason is that Social Security isn’t a safety-net.

This statement by the Trustees is easily the most misunderstood in the entire Social Security debate. The media has misread what the Trustees have said, and then place their translation well out of context. Readers and bloggers subsequently surround that misconception with words like guarantee which do not even appear in the original text.

This statement is part of a much larger report which is a cautionary tome. The Trustees are providing this guidance to the public as a reasoned warning. “This year’s Trustees Report contains troubling, but not unexpected, projections about Social Security’s finances. It once again emphasizes that Congress needs to act to ensure the long-term solvency of this important program.” They are warning people who are as old as 63 that they can expect to be affected by the problem in the system.

In contrast, some pundits have taken the guidance out of the cautionary context, and replace it in an upbeat tale of success. Paul Krugman has reframed this warning as “It’s also worth noting that even if the trust fund is exhausted and no other financing provided, Social Security will be able to pay about three-quarters of scheduled benefits”. The guidance is transformed from a caution to a measure of success.

What does the Trustees statement really mean? Once the Trust Funds are exhausted, benefits cannot exceed the inflow of the system. Since inflows will vary from week to week and month to month, one has to assume that benefits will likewise vary. One week you may receive 80% of your scheduled benefits, and 70% a different week. In the statement, the Trustees project that Social Security will distribute to all recipients on average 75% of scheduled benefits over the course of a year.

This statement does not mean that an individual’s benefit check will be reduced by 25%. What the system can pay in aggregate is completely unrelated to what you can expect to receive as an individual. There is no law which allocates the reduction caused by insufficient payroll tax collections. According to IBD, public trustee Charles Blahous has said that low-income and elderly beneficiaries would have to be protected from cuts, meaning the possibility of a bigger impact on others . If so, your check might be 100% or 50% of your scheduled benefits.

2033 is not a hard date nor is 75% of scheduled benefits a hard number. These are projections which the Trustees have provided to demonstrate to Congress the seriousness of the problem. Both of these numbers are estimates based on assumptions which may or may not materialize. In fact, the Trustees have provides scenarios in which the Trust Fund is exhausted as early as 2027 . The exhaustion date has changed over the past few years as previous assumptions have proven overly optimistic. 2033 assumes a good economy. If the economy is less than good, Social Security will run out of resources sooner.

One of the assumptions that enable Social Security to meet 73% of its obligations in 2086 is that workers will continue to pay 10.6% of wages in payroll taxes for retirement benefits . This assumption is optimistic. Social Security isn’t just retirement insurance. It also provides parental support costs. Once the Trust Fund is exhausted, workers will be faced with a system that doesn’t provide the same level of insurance at a time when they are paying more to support their parents. So the Trustees are assuming that the working generation will pay the same price for a system which delivers less at a time when disposable income is going down. There is virtually no economic model that explains that assumption.

Workers today enjoy a holiday from payroll taxes. It started in 2011, and was continued in 2012. The problem with ending this holiday is that politicians have an exceedingly difficult time explaining to 160 million voters why their payroll taxes should return to the former levels. If you can’t explain increasing taxes today, imagine trying to explain maintaining payroll taxes to someone who is losing not only personal benefits, but paying incremental support for their parents because the system has failed.

In the net, the Trustees do not know when the system will lose the capacity to pay full benefits nor can it tell you how much you will receive once the Trust Fund is exhausted. You cannot plan when your benefits will be reduced or by how much. The guidance is a warning about the problem in the system, not a testament to its success.

The one thing that is missing from the Social Security debate is a discussion about the assumptions on which the Trustees have projected the soundness of the system.  Today's debate reduces hundreds of pages of data into 4 digits - 2033 in this case - without asking a single question about the assumptions that went into creating that number. 

2033 is only as good as the underlying assumptions. If the Trustees say that the system will last until 2033 based on the idea that leprechauns will spit-out gold coins to pay benefits, no one should be surprised if the system collapses long before 2033. While many people quote the 2033, no one is questioning the assumptions - some of which are very generous.

The most generous assumption is that Social Security is a closed-system, unaffect by the outside world.  For example, the Trustees Report ignores the impact of the pending insolvency of the Medicare Trust Fund. According to the Social Security Administration, the calculation of the OASDI exhaustion date does not reflect any impact of potential exhaustion of the Medicare trust funds, as the two programs are separate entities.

Social Security does not exist in a vacuum as the Trustees portray. Social Security consists of three 'separate' systems, Old-Age/Survivors(OAS), Disability(DI), and Medicare(HI). They aren't separate - all three draw on the same tax base - payroll taxes.  So anything that affects payroll taxes will affect Social Security's OAS program.

The Trustees project that Medicare will reach insolvency in 2024.  At that time, Congress will have three choices.

1. Pull payroll tax resources away from OASDI – leading to an earlier date of exhaustion in OASDI

2. Pull general tax resources away from deficit control or debt reduction – leading to higher levels of debt

3. Redefine Medicare benefits.

Social Security will not be immune to this debate or its outcome.

Another example of an unrealistic assumption is the impact of cutting Social Security benefits on payroll taxes.  As benefits to seniors drop by 25%, the spending of seniors will drop.  That will lead to lower GDP, on which payroll taxes depend.  As aggregate spending drops so will payroll taxes and so will payments to seniors.

On the other hand, the Trustees may be right.  They are assuming that aggregate spending will not be affected by reduced benefit structure.  They assume that seniors will continue to spend despite lower benefit checks. If so, Social Security should be able to provide roughly 75% of scheduled benefits until 2086.

Some will suggest that it isn't possible to incorporate behavior responses to exogenous changes.  But the fact is that the Trustees do selectively include foreseeable events.  For example, the Trustees tell you about the impact of insolvency in disability - just not in Medicare.  For example, the Trustees' forecast anticipates lower income tax revenue (which is rebated to SSA) - just not lower income from a drop in spending. 

This article isn't to suggest that the Trustee's have failed in their duty.  The point is to provide clarity to the Social Security reform debate.  When you hear "Social Security will take in enough revenue to keep all of its promises for over 30 years, without any changes at all", you know that is provided that leprechauns spit-out gold coins.

Footnotes :

Anyone who thinks that medicare and OASDI are unrelated should look at Appendix F “Estimates For OASDI and HI, Separate And Combined” discusses the cashflow of the combined entities.

In 1983, Social Security was insolvent.  Every penny ever collected for Social Security had been distributed to beneficiaries.  The revenue collected from payroll taxes was insufficient to cover the annual benefits of current recipients.  Social Security was forced to borrow money so it could pay the promised benefits – and the politicians went to work.

Most of the reform was based on the work of The Greenspan Commission which was formed in 1981.  It was chaired by Alan Greenspan, the man who would overtime bring us the housing crash.  The recommendations consisted of raising taxes and cutting benefits.   The politicians applauded the hard-fought compromise, and told the country that the system was fixed.

Like other fixes, the fix was short-lived.  The system now has 20.5 trillion dollars of unfunded liabilities¹.  That means that the system has roughly $10 of promises for every dollar of asset.  One can see clearly that the system isn’t fixed at this time.  One can argue that it wasn’t fixed in 1983. 

Understanding this failure is essential for anyone who hopes to save the system.  The failure is actually very simple to understand.  The country pushed the legacy costs of Social Security disproportionally on to non-voters with the assumption that they would cover the increasing costs.  Non-voters received substantially larger benefit cuts and substantially larger cost increases.  Now these people can vote, and there is no way to bind them to the terms of the 1983 agreement.

Over time, these people who had no vote in 1983 have grown into a massive voting block.  In fact, 2010 was the first year in which a majority of voting aged-Americans could expect to have benefits reduced as the Trust Fund runs dry.   This voting block results from two factors.  People who were non-voters have gotten older.  In 1983 someone who was 17 is now 46.  The other factor is that the promises of DC are proving optimistic.  Over the last 5 years, Run-Dry Date has dropped by 8 years.  As that run-dry date gets closer, it will affect more and more Americans, and they will vote accordingly.

Summary Of The Greenspan Commission

The people who were non-voters at the time of the 1983 changes could expect to pay the higher rates of taxes than any generation and were subjected to larger cuts in benefits. 

Increased tax rates :

“Advances scheduled increases in Social Security tax rates. Social Security tax rates (which include the Hospital Insurance tax rates) for employers and employees will increase to 7.0 percent in 1984, 7.05 percent in 1985, 7.15 percent in 1986-87, 7.51 percent in 1988-89 and 7.65 percent in 1990 and thereafter.”

Today Social Security’s portion of payroll taxes is 10.6% of the 15.3%².  If you started work in 1990, you expected to face 49 years of peak rates.  If you were 40 in 1983, you could expect to face 26 years of peak rates.  If you were 45 in 1983, you could expect to face 20 years of peak rates.

Adjustments to retirement age:

“Raises the age of eligibility for unreduced retirement benefits in two stages to 67 by the year 2027. Workers born in 1938 will be the first group affected by the gradual increase. Benefits will still be available at age 62, but with greater reduction.”

In 1984, the retirement age of someone who was 45 was unaffected.  Someone born in 1938, faced a modest increase in retirement age.  Someone born in 1960 and later got the full increase of two more years of work.  So the majority of the savings comes at the expense of non-voters in 1983.   

Introduction of means-testing :

“Beginning in 1984, includes up to one-half of Social Security benefits as taxable income for taxpayers whose adjusted gross income, combined with half their benefits and any tax-exempt interest they may have exceeds $25,000 for a single taxpayer and $32,000 for married taxpayers filing jointly. Benefits received by married taxpayers filing separately are taxable without regard to other income. Appropriates amounts equal to estimated tax liability to the Social Security trust funds.”

“Changes the earnings test for beneficiaries age 65 and over so that $1 in benefits will be withheld for each $3 of earnings above the annual exempt amount, beginning in 1990”

In 1984, this rule did not affect many people.  The problem is that these limits have not been changed for inflation.  These rules also have a greater impact on people who saved for retirement with IRAs or 401Ks.  Today it affects up to 1/3^rd of retired Americans.

The Lesson Of 1983

The lesson is simple.  You can't solve the legacy burden of Social Security by voting to put the costs on non-voters.  Overtime, these people will grow into voters that will not honor the terms of the agreement.  By sheltering one voting block at another’s expense, we opened Pandora’s Box.  Every generation will feel entitled to shift the costs that were given to them to the next generation.  When a generation says 'no', it will create a very difficult transition.

1, 2012 Social Security Trustees Report Page 15

2. The payroll tax holiday reduces OAS rates from 10.6 to 8.6% on a personal basis.  It raises an offseting amount from the general taxpayer.  So FICA really mains at 15.3% of wages.

One of the talking points in DC today says that Social Security has not added one penny to the federal government’s  budget deficit.  In its most extreme form, you will hear that Social Security has actually reduced the deficit.  When you hear it, keep one fact in mind: in politics, the number of beans is never as important as how you count them.

Only in DC can you spend ½ a trillion dollar every year, and it is not part of your deficit.  Better yet, only in DC can you spend ½ a trillion dollars every year, and it cuts your deficit.  It is like when I tell my wife that the cheese cake I am eating actually burns calories as I move the fork to my mouth.  Yes, it is absurd.

The politicians say it because voters want to hear it, just like you want to hear about a weight control program that consists of cheese cake.  In the minds of these politicians, Social Security is a closed system which operates in a world in which people like taxes.  In such a world, Social Security would not contribute to the deficit.

The problem is of course that Social Security is not a closed system and people do not like taxes.  Payroll taxes affect jobs.  In fact, they kill jobs.  According to the Congressional Budget Office, lowering payroll taxes from 15.3% to 13.3% will create as many as 7 million jobs.   Social Security portion of payroll taxes is 10.6%, not 2%.  So Social Security costs millions of jobs and billions in income tax revenue.  Of course, in the mind of politicians, Social Security taxes do not affect jobs.

Payroll taxes increase cost of hiring Americans by 15.3% which makes our goods less competitive in world markets.  In my case, I was a programmer in a world where the government added 15.3% to my labor cost but didn’t add it to imported labor.  Technology is a very labor intensive business.  So a major part of that industry is now offshore.  But in the world of the politicians there are no job losses.

Payroll taxes decrease the incentive to work overtime or work longer in your career.  Payroll taxes reduce your take home pay by 15.3%.  Payroll taxes can reduce your take home by more than $15,000.  In the case of older workers, this cost will have little impact on their benefits.    So many older workers retire early who value their time.   The problem is that this group is apt to be concentrated in high-income earners who would be paying income taxes.  But in the world of the politicians, no one retires early.

Payroll taxes also affect income taxes.  As payroll taxes go up, income taxes have to come down, unless people like taxes.  There is a maximum amount of taxation that the economy can carry.  When one tax goes up, others have to come down in order to maintain the balance.  Historically 19.5% of GDP is roughly the maximum.  When payroll taxes are 15.3%, it leaves very little left over to collect for low-wage workers to pay income taxes.  This is why people who earn the top 51% American households have no income tax obligation.  Basically we are paying into our retirement accounts, and putting the rest of the government on the credit card.  I am not sure how that doesn’t add to the deficit in the minds of politicians. 

Milton Friedman once observed, “If the government ran the Sahara, in five years there would be a shortage of sand.”  He is right.  The government has run Social Security for 70 years, and now there is a shortage of security.   Today more than 80% of Americans believe that Social Security is heading for crisis if the government does not implement a major reform.  So a system that is supposed to provide security, now it only provides uncertainty for the vast majority of Americans. 

Friedman’s quip about the Sahara is an amusing look at government.  The comment about Social Security on the other hand is rather frightening.  The difference between these similar outcomes is that few people depend upon sand in the Sahara; where as Social Security has become a sinkhole of dependence with millions dependent upon a system which is comically broken.

Just how broken is the Social Security system?  While people argue about the solvency of the system, the Social Security Trust Fund’s assets are managed with a 70 year-old investment policy that has underperformed the equity markets by nearly 50 to 1 during that time.  As a consequence, we are debating raising taxes which will be subsequently invested in bonds with a yield of less than 3%.

One day, economists will study this comedy because Social Security is the perfect storm of economics.  It blends the inefficiency of monopolies, with the incompetence of government, with the indifference of absolute pricing power.  This concoction isn’t just headed for failure.  It is headed for massive failure.

People tend to think of the Post Office as the poster-child of government ineptitude, but it is well run compared to Social Security.  The reason for its relative success is that the Post Office has to compete with private sector companies.  No one is forced to use the United States Postal Service.  We communicate by phone, email, and best of all private sector mail services which drive innovation into the business model of USPS.

Social Security competes with no one.  It could be the purest monopoly in the world.  While thousands of firms offer investment products, Social Security does not compete with any of them.  Social Security gets more than half a trillion dollars every year regardless of what happens in the outside investment world. 

As a consequence, innovation is driven by managers at the Social Security Administration rather than by market needs.  The Social Security Administration chose to add things like automated check deposit.  But it didn’t have to offer that service.  The Social Security Administration offers a website, but it didn’t have to offer one. 

So how bad is letting government drive innovation?  Only the government could create a retirement tool that is completely insensitive to risk appetite.  Social Security allocates risk in a one-size-fits-all model that is not only expensive but dangerous.  Imagine, the nation is engaged in a discussion about the solvency of a system, which has no way to allocate the risk associated with insolvency.

Can government innovation get worse?  Social Security is supposed to provide insurance, and yet it has no mechanism for price discrimination.  Survivor benefits are difficult to price – so we make them free.  The unhealthy cardiologist with a family of four pays exactly the same rates that a single person does.   According to the Social Security Administration, survivor benefits basically double the cost of benefits - but the benefits are given away for free.

Actually it can be worse still.  The Social Security system cannot invest in higher yielding assets because some critics feel that the financial markets are too risky.  As a consequence, Social Security invests 100% of its excess assets in a single issuer within an asset class that has enjoyed a 30 year bull market.  The issuer is the US government which has Debt/Annual Revenue of more than 500%.  But the financial markets are too risky.

In short, Social Security could be the only retirement product in the world that cannot allocate risk.  It is likely the only insurance product in the world that can’t price risk.  And it has an investment strategy which brings together low returns and maximum risk.  Social Security is broken because there is no private market where innovators can drive incompetence out of business.

In the private sector, profit and loss regulate the decisions of such managers.  Government has no way to incorporate profit and loss into its decision making process.  In the mind of its managers, positive cash flow is profit regardless of what happens to unfunded liabilities.  This is not a joke.   Politicians of both parties generally agree that Social Security has not contributed to the deficit, ie it makes money.   So Social Security will not be a problem until it is a catastrophe.

The consequence of absolute pricing is that no one who runs the system cares about whether the product is any good.  If the system runs out of money, the managers simply raise the price and shrink the box.  In terms of Social Security, raise the price and shrink the box means raising taxes and lowering benefits.  In the last 70 years, we have never had a single discussion about how to fix the system.

Today some actually argue that we should raise taxes and lower benefits because “Social Security is the most successful government program ever.”  It would be funny if millions of people did not depend upon the system.  As a system, Social Security does not attract money well.  It does not manage what resources it has well.  And it does not allocate its resources well to serve its purpose.  Social Security is horribly broken.  Our leaders don’t want to fix a broken system they want to convince us to pay for one.

That is not funny.

The budget deficit is the greatest threat to Social Security.  As deficit increases, interest costs create an increasing footprint on our tax base.  As interest costs go up, the ability of workers to pay into Social Security goes down.   

Some think that the problem with Social Security is the number of workers to retirees.  The flaw here is that workers do more than support retirees; specifically they pay income taxes which support the deficit.  Today, the average worker is carrying almost $100,000 of look-through debt from the federal government alone.  The additional debt burden means that the worker today is not comparable with the worker of 2000.

We are kidding ourselves if we believe that Social Security doesn’t add to the deficit.  Senator Sanders is going to introduce a bill to increase payroll taxes.  If we can raise taxes, we should raise taxes to lower the deficit, rather than to fund a retirement account.

I am a boomer, and we have left our children a staggering debt.  We shouldn’t be surprised if they resent the obligation to take care of our debt and the people who created it.