Why Is the Chained CPI in the President’s Budget?

My last post took on some of the basic questions raised by the debate over using the chained CPI (Consumer Price Index) to adjust Social Security benefits.

I deferred the question in the headline here because the post was already quite long, and the answer isn’t simple. So here goes …

Social Security and the Deficit

Strictly speaking, Social Security doesn’t belong in the budget at all — at least, not in the package of spending and revenue proposals we ordinarily think of as such.

It has its own revenue stream — the payroll tax, plus an earmarked portion of income taxes paid on some of the benefits it provides. It also has $2.7 trillion in reserves, i.e., the unused portion of these taxes, invested in Treasury bonds, and the interest on these.

That’s all it’s got.

A shortfall would be dreadful, but it would have no impact on the deficit — unless, as seems likely, Congress used general tax revenues to avert a sudden, big benefits cut.

This, however, is an argument for crafting a measure that will keep the program solvent, not for putting the chained CPI in the budget.

Some say that the Trust Fund is just an accounting fiction. The Treasury bonds the reserves are invested in signify money that’s being used to help pay for items in what we ordinarily think of as the budget.

When Social Security starts drawing on its reserves, as it already has, the Treasury Department has to sell some bonds to other investors in order to pay the program what it owes — or use revenues from taxes not specifically intended for Social Security.

This doesn’t mean that Social Security is contributing to the deficit, however — any more than you or I could be said to increase the deficit if we cashed in some savings bonds a grandparent once gave us.

More Revenues Without Tax Reform

The chained CPI is probably in the President’s budget in part because it would increase tax revenues without any rate-raising or loophole-closing at all.

According to Congressional Budget Office estimates, the federal government would gain $123.7 billion* over the first 10 years because tax brackets and other annually-adjusted tax provisions, e.g., the personal exemption and standard deduction, would rise more slowly.

So even a quite small increase in income could get taxed at a higher rate. The amount we’d owe wouldn’t be a whole lot greater than what we’d owe if the Internal Revenue Service continued to use the same inflation measure it’s been using.

But the tax code would be somewhat less progressive because filers at fairly low and moderate-income levels would take the biggest hits.

And at least some low and moderate-income families would get smaller reductions and/or refunds from the Earned Income Tax Credit because the maximum credit is adjusted for inflation, as are the phase-outs that gradually lower the credit when earnings reach some level above the amount eligible for the maximum.

There would be no impact on the refundable Child Tax Credit if Congress makes the current threshold for claiming it permanent, as the President has proposed.

Big if here, since we know that Congressional Republicans have wanted the EITC and the Child Tax Credit to revert to their more restrictive pre-Recovery Act forms.

Political Strategy

The revenues raised would be a small portion of the total increase the President now says he’d settle for. So it’s pretty clear the chained CPI is in the budget mainly for strategic reasons.

The received wisdom seems to be that he’s again striving for a grand bargain — offering Republican Congressional leaders the chained CPI and Medicare spending cuts they said they wanted in the fond hope they’ll agree to a scaled-back revenue-raising plan.

Or if not that, perhaps proving they’re altogether unreasonable and ought to lose their House majority next year so that Congress can get important business done.

This is what Michael Tomasky at The Daily Beast thinks the President is up to — and why he thinks no one should fret about the chained CPI.

The Nation‘s John Nichols thinks otherwise. Look, he says, at how the chairman of the National Republican Congressional Committee is already messaging the President’s proposal as a “shocking assault on seniors.”

This is likely to depress votes for Democrats next year, Nichols predicts, citing examples from past mid-term Congressional elections.

The chained CPI proposal certainly has complicated life for Democrats in Congress now, even if they ultimately don’t have to cast an up-or-down vote on it — still TBD.

The larger issue, I think, is that the President has, to some extent, legitimized use of the chained CPI as a way to “save” Social Security — and chosen it instead of lifting the payroll tax cap.

So, as Blake Zeff at Salon asks, “How hard would it be for Republicans to push cuts through, when this [the chained CPI] is now mainstream Democratic policy?”

Cuts, I’d add, that could extend to programs specifically for low-income people, which the President’s proposal would hold harmless.

Note how House Majority Leader John Boehner grudgingly welcomes the chained CPI as an acknowledgment that “our safety net programs are unsustainable.”

This implies something far more sweeping than what the President has proposed for Social Security and Medicare, which arguably aren’t safety net programs anyway.

Well, maybe the rumblings and grumblings, mine included, are just worst-case scenarios. But I’m not ready to bet on that.

* The Office of Management and Budget estimates the revenue gain at $100 billion. Differences between CBO and OMB estimates are not unusual.

Chained CPI and Social Security: Some Questions and Answers

Back in January I said I’d delve into the impacts of using the chained CPI (Consumer Price Index) to adjust Social Security benefits.

Then the President delivered a strong defense of “the commitments we make to each other,” including Social Security. So I put my draft post aside, figuring the chained CPI was off the bargaining table.

The more fool I. As you’ve probably read, it’s among the entitlement “reforms” in the President’s proposed budget.

The White House has taken great care to package it with other changes that would supposedly protect very elderly retirees and others who’ve relied on Social Security benefits for a long time. Also to shield programs that base eligibility on income.

But as economist/blogger Jared Bernstein said some time ago, the danger is that protections like these will all get swept aside, leaving only the benefits cuts.

Here then are some of the basic issues, as I see them. More to follow in a second post.

Why Do Anything About Social Security?

Our federal policymakers must do something about the Social Security retirement and disability insurance programs — and the sooner the better.

The Trust Fund will run out of money long about 2033. If nothing is done before then, the Social Security Administration will have to rely solely on what it continuously receives from payroll taxes.

That would be enough to pay about 75% of the benefits retirees would get if the Trust Fund still had reserves — a devastating loss for the nearly two-thirds who rely mainly or entirely on those benefits.

What Could Prevent the Shortfall?

At the risk of over-simplifying, our policymakers have two choices, not counting just letting the Trust Fund dry up. They could change the system to take in more or pay out less.

A switch to the chained CPI represents the latter approach because it rises more slowly than the currently-used index — the CPI-W. So, therefore, would the benefits that seniors, severely disabled former workers and eligible survivors receive.

Policymakers could instead “scrap the cap” on the amount of wage income subject to the tax that feeds the Social Security programs.

Or they could raise it enough to cover all but the top 10% of income, as it did in 1982 after Congress stepped in to shore up the program — not for the last time, incidentally.

High earners wouldn’t like this, of course. Nor perhaps would employers, since they’re responsible for half the payroll tax.

The National Federation of Independent Business has already said that its small business members “would violently oppose” it. These, as you know, are the “job creators” that our President and Congressional leaders are so fond of.

Is the Chained CPI More Accurate for Cost-of-Living Adjustments

Proponents of the chained CPI claim that it’s simply a more accurate cost-of-living measure because it reflects consumer responses to price increases. If the price of beef goes up, they buy less of it and more chicken. Etc.

Opponents argue that the index isn’t more accurate for seniors because they spend far higher portions of their income on items that aren’t amenable to switches, especially health care.

Even the CPI-W apparently understates their cost-of-living increases.

For some time now, the Bureau of Labor Statistics has maintained an experimental cost-of-living index specifically for the elderly. Over the long haul, it has risen somewhat faster than the CPI-W. And Social Security’s chief actuary expects it will in the future.

We’d probably see similar results from a cost-of-living measure for people with severe disabilities, since many of them also have disproportionately high health care costs.

So if accuracy were the real issue, Congress would give BLS the funds to fully develop its experimental index, as The New York Times, among others, has suggested.

What Other Objections Have Opponents Raised?

The over-riding objection to the chained CPI switch is that it would effectively cut benefits. The loss in any one year would be small. But losses would mount up over time because the base for each cost-of-living adjustment, as well as the COLA itself, would be lower.

The average earner, says the Strengthen Social Security Campaign, would lose a total of $4,631 by age 75 — more than three months of benefits. Another 10 years and the loss would mount to nearly a year’s worth of benefits.

We need to recall that retired workers now get, on average, only $1,261 a month — and former workers in the Social Security Disability Insurance program somewhat less.

For 36% of seniors, Social Security provides at least 90% of income — not surprising, given what we know about retirement savings. It’s the sole source of income for 29% of the most elderly.

As I mentioned earlier, the President’s budget includes a “benefit enhancement” — popularly known as a bump up. It’s supposed to restore the cumulative losses for people who live long enough to benefit.

Most who do would still get less, according to the SSS Campaign.

Those who don’t are just out of luck, of course. And they’re disproportionately lower-income people, for whom every Social Security dollar counts.

UPDATE: After I published this, the Center on Budget and Policy Priorities issued a brief that provides more detail on the impacts of the chained CPI on Social Security retirement benefits.

“Ponzi Scheme” Keeps Nearly 14 Million Seniors Out Of Poverty

I’ve remarked before that the measure the Census Bureau uses for its annual poverty reports was crude from the get-go. Three times the cost of what was then the U.S. Department of Agriculture’s lowest-cost food plan.

The measure has become increasingly inappropriate over time, as food costs have come to represent a smaller portion of basic living costs and living standards have risen.

Shawn Fremstad at the Center on Economic and Policy Research tells us that, in the early 1960s, the poverty line was nearly half the median average income. It’s now slightly under one-third.

Other research, he says, suggests that twice the applicable poverty threshold would be a better indicator. That would put 33.9% of Americans, rather than 15.1%, into the poverty category.

Well, the Census Bureau will be issuing poverty figures based on an alternative measure it’s developing. In the interim, it gives us a bit of a preview, with three “alternative resource measures” pulled from its 2010 Income, Poverty and Health Insurance Coverage report.

One shows how poverty figures would rise if the current measure didn’t include Social Security benefits. Hence the headline.

In 2010, these benefits kept a total of 20.3 million Americans from falling below the poverty threshold. The largest number — 13.8 million — were 65 and older.

In other words, without Social Security benefits, five times as many seniors would have been officially poor — and therefore very poor indeed.

Specifically, a senior living alone would have had to get along on no more than $10,458 for the year, a couple on no more than $13,194.

To put this in perspective, an elderly couple in the Washington, D.C. area would have had only $678 left after paying the rent for a modest efficiency unit.

The other alternative measures show similar, if somewhat less dramatic impacts. For example, 3.2 million people had enough income to lift them above the poverty threshold because at least one household member received unemployment insurance benefits.

The current poverty measure includes Social Security and UI benefits because they’re cash income. Food stamp benefits aren’t.

But, as the Center on Budget and Policy Priorities reports, 3.9 million fewer people would have been counted as poor if they were part of the Census measure.

These figures ought to send a message to policymakers who want to bring the long-term deficit down by drastic cuts in Social Security, as well as key safety net programs.

Also, I think, provide useful perspectives on what they’re hearing from some conservative economists and columnists.

Professor Edward Glaser, for example, argues that the temporary payroll tax cuts in the President’s jobs plan should be paid for by increasing the eligibility age for Social Security retirement benefits because that “was always going to be part of any sensible entitlement reform.”

So much for those with other ideas.

Washington Post columnist Charles Krauthammer puts a new spin on his long-standing insistence that Social Security (and Medicare) benefits have to be cut because too many of us older folks are living too long — and didn’t have enough kids when we were younger.

Social Security is a Ponzi scheme, he says, implicitly endorsing the view of that well-known economist Governor Rick Perry.

This, I take it, is because Social Security operates like other insurance programs, with premiums some people pay, plus returns on investments, funding payouts on claims by others.

Still, Krauthammer allows that Social Security is “vital” and “humane.” Just needs to be adapted to “new demographics.”

Most important fix is to raise the retirement age — “an absurd anachronism” he calls it, using life expectancy figures that the Social Security Administration itself has shown to be misleading.

This isn’t to say that the Social Security Trust Fund won’t eventually lack funds to pay full benefits if we do nothing to change the system. Only that a further increase in the retirement age is hardly the only way to fix it, let alone the best.

What the new Census figures tell us is that a further hike in the retirement age will almost certainly increase the over-65 poverty rate.

What, pray tell, are older workers supposed to do if they lose their jobs or find the tasks too physically demanding?

Just hang in there in until they’re 70? Take a big cut in benefits at a bumped-up early retirement age — a likely companion to an increase in the full retirement age?

Why not instead address the future shortfall by lifting the cap on the payroll tax? Or at least, as Senator Barney Frank (I-VT) proposes, requiring the highest-earning workers to pay more?

A Second Look At Flat-Lined Social Security Benefits For Seniors

Before I decided to publicly reject the administration’s proposed $250 Social Security gift, I thought about potential cost-of-living increases that could disproportionately affect retirees. Turns out I gave up too soon.

I knew all too well that Medicare premiums deducted from Social Security retirement benefits can, over time, erode their value because health care costs are rising faster than the overall inflation rate.

But I read that nearly 90% of beneficiaries would be protected from a bigger premium bite next year. Of these, 70% are covered by a “hold harmless” provision, which says that beneficiaries already paying Part B premiums won’t get tapped for increases that would reduce their net benefits, unless they earn enough to subject them to income-adjusted premiums. The remaining 20% because state medical assistance agencies will pay the premiums.

What I’ve since learned is that there’s no “hold harmless” provision for Medicare Parts C and D — the optional alternative coverage offered by private insurance companies and the also-optional insurance coverage for prescription drugs. So some portion of that 70% may have less for other living expenses.

More importantly, I’ve found compelling evidence that the cost-of-living index used to determine the need for benefits adjustments — the Consumer Price Index for Urban Wage Earners and Clerical Workers — doesn’t accurately account for seniors’ rising living costs.

Higher health care costs are a major factor here. Others include higher housing costs and, to a lesser extent, fuel oil costs. We know this because the Bureau of Labor Statistics has been maintaining an “experimental” Consumer Price Index for Elderly Americans since 1982.

It’s found that, over a 25-year period, the CPI-E has risen somewhat faster than the CPI-W used to calculate benefits. The long-term differences translate into an average 3% difference a year. The difference doesn’t matter much in the first year, but it mounts up over time.

For example, according to a recent report by the Senior Citizen’s League, someone who received $816 in 2000 would have received $216 less in 2009 than what the CPI-E indicates would have been enough to maintain a steady state.

Now the CPI-E is only suggestive. As BLS explains, the sample used to construct it is small. And it’s not an altogether accurate reflection of either the demographics of Social Security retirement beneficiaries or their living costs.

It nevertheless suggests that cost-of-living adjustments in retirement benefits ought to be based on a more suitable index.

Congressman John Duncan, Jr. (D-TN) has introduced a bill — the CPI for Seniors Act — that would direct BLS to develop and publish a monthly index of changes in consumption expenditures that are typical for people aged 62 and older.

It’s got five cosponsors and probably not going anywhere fast. Because once you’ve got the index, you’ve got solid grounds for bigger annual increases — thus greater stress on the Social Security Trust Fund we’re supposed to be so worried about.

And you’d still need another index to account for Medicare and other expenditures that are singular or different for the over-full retirement age population.

None of this, to my mind, argues for the proposed $250 across-the-board consolation prize for no cost-of-living increase next year. Nor against my view that the $250 would be better spent as assistance to struggling TANF families.

But it does say that I shouldn’t have been so quick to dismiss the complaints about the unfairness of another year without a COLA.