Every once in awhile Congress and the President come together and are able to come up with solutions to problems. This makes me suspicious. I can count on one hand (finger maybe) the number of truly positive, trans-formative bills that have passed with bipartisan cooperation over the last 40 years. Typically any bill worth its salt has had to pass with bare majorities, parliamentary tricks, or over a glass of Merlot in a quiet corner.
The last 4.5 years have been no exception, with the Republicans acting as a human Maginot Line against Democratic policy. Very little policy of substance has passed Congress since 2010. In the meantime while employment has languished, our credit rating cut, our economy plows ahead dragged by the private sector the Fed, and some states. Congress lies broken. It’s fallen and can’t get up, it’s enormous fiscal power put away in a secure room where no one but the well-connected can benefit from it.
Lately however there seems to be some growing consensus around one idea to help our deficit and debt issue. The President and some Democrats and Republicans seem willing to consider changing how the inflation calculation for Social Security and other federal benefit recipients are calculated. Republicans seem willing to listen and even seem willing to exchange other items in return.
The reaction from citizen’s groups, the media and others has been diverse. Some see this as a “Grand Betrayal” to seniors and future SS recipients. Others see this as a much-needed fix to an important social insurance program, still others see this as a tool to finally break open the ice that grips Congress and the President.
Let’s leave the politics aside for a moment and examine what “Chained CPI” is.
Then let’s examine the benefits and drawbacks of applying it to Social Security. ( I know it will be applied to other federal programs as well, but we’ll stick with SS for this article)
Then we’ll circle back to the political landscape and see if, by understanding Chained CPI and Social Security better, we’ve resolved not only whether it’s a “good” idea, but whether or not it’s necessary.
What is it? -
Many years ago the federal government admitted that stuff tends to get more expensive every year. They realized that that general insurance programs like Social Security and more specific programs like veteran’s benefits and federal employee pensions should increase over time to reflect these increases in costs. However, different things get more expensive differently. Over time, more traditional standard measurements of inflation became broken down into ever more focused descriptors.
For example the Consumer Price Index (CPI) is a measurement of a fixed basket of goods and services that the average American might buy or use. This “average basket” of items can change and often does over time but is typcially comprised of “Household Items, Personal Goods and Services, Tobacco, Leisure Goods, Households Services, Housing, Alcoholic Drinks and other categories”.
But some of these items are more difficult to measure than others. For example food and energy prices can rise and fall very rapidly and unpredictably. This can pose challenges to government agencies that are trying to calculate what increases to embed in Social Security of other federal benefits.
As a result, most statistical agencies use a ‘Core Inflation‘ measurement that removes food and energy prices from the calculations, feeling it gives a more accurate picture of what “real” inflation looks like. The Fed , for example, used this particular measurement to give it the most accurate picture of how they should be reacting to inflation (up until 2012).
There are thousands of subsets of the CPI. 8,000 as a matter of fact. Each one is tracked and recorded by the Bureau of Labor and Statistics. The BLS tracks the prices of 211 consumption items in 38 geographic areas. For our purposes we really only need to be introduced to 3 main actors:
Social Security currently uses “CPI-W“, which is the Consumer Price Index for Urban Wage Earners and Clerical Workers. This “basket of goods” that would typically be consumed by the typical urban worker that includes clerical workers, sales workers, craft workers, operative, service workers, or laborers. Specifically excluded from this measurement are professional, managerial, and technical workers; the self-employed; short-term workers; the unemployed; and retirees and others not in the labor force. More than one half of these “consumer units” as they’re called (the workers and what they consume being measured) must come from one of the above occupations. This particular index has its historical roots going back to WW1, when it was created to help in labor negotiations.
“CPI-U” – or – The Consumer Price Index for All Urban Consumers was created in 1978 to measure urban consumption because it measures the consumption of all urban workers rather than a particular subset. Rural workers, and the institutional population are excluded from this measurement.
“CPI-E” – or – The Consumer Price Index for the Elderly was created in 1982 to measure the buying habits of the elderly which are typically much different than the general population. For BLS purposes, once someone (or their spouse) turns 62, they meet the definition of “elderly”. Individuals in this group typically consume twice a much medical goods and services than people measured in our other two indexes. The CPI-E traditionally has outpaced the CPI-W and CPI-U by .20% a year on average, reflecting the rapid increase in medical costs in the US for seniors.
Stick with me for just a bit more, I promise the edge of woods is close….
Indexes are comprised of the population and goods they buy factored with a set of “weights” to make them more accurate. Housing for example, is the most important product or service bought by urban consumers, so housing is given a much higher proportion of the total calculation of the index. For example housing might comprise 40% of the CPI, while Transportation might make up 15%, Alcohol 5%, etc.
Here’s why that’s important: The prices of all of these goods and services are constantly changing. The BLS has to adjust the weighting in all of their indexes to reflect this. They adjust the weighting in the CPI-W and CPI-U every January of an even-numbered year. This isn’t very often given the volatility of the prices and buying habits of consumers.
“Chained CPI-U” adjusts the weighting monthly. Many economists argue that this makes the index a more accurate reflection of consumer habits and prices. The BLS certainly does, or it wouldn’t have invented it specifically for this purpose.
Using Chained CPI-U methodology results in a measure of inflation that typically runs .25-.30% lower than CPI-W or CPI-U. Everyone and their mother knows this now. But why? Why is it lower? The BLS has always calculated in their weights that if the price of red apples increased, then people would switch to lower-priced green apples. What they didn’t figure *(until 2002) is what if people simply switched to oranges? What this does is that big price spikes in some goods and services are muted in the Chained CPI because consumers will often switch to a completely different product or service before that price spike is finished. This effectively “smooths out” the biggest swings in the volatility of the CPI, which results in a slightly lower measure of price increases and therefore inflation.
Whether or not this explanation is clear to you, one thing is clear: The BLS measures the shit out of everything.
If you are still awake, let’s move forward.
What does it mean for Social Security? And what is the President’s specific proposal?
So why is the index used to determine COLA’s for Social Security even being considered? Leave out the deficit fetishists, they aren’t even up for serious discussion here, but many other friends of Social Security are thoughtfully considering a change to this index? The reason? Social Security does have some minor problems.
While the Social Security Trust Fund is still growing, under current law and funding levels, the huge influx of baby boomers moving through the system will eventually stress the system until it is only able to pay 75% of scheduled benefits starting in 2033. No one wants to see that happen. With only 20 years or so until it does, making some minor adjustments to Social Security, its funding or its benefits is mandatory.
If Chained CPI-U is adopted as the index used to adjust future Social Security payments, the CBO estimates that it would mean a decrease payments from the Social Security program in the neighborhood of $130 billion. It also estimates these changes and the changes to other federal program inflation calculations and savings on interest on the national debt would create a total of $390 billion in deficit savings. For our purposes today lets assume these numbers are accurate (because we know by definition that every forecast is inaccurate but going down that rabbit hole is simply boring).
The bottom 60% of recipients would see a 3% lifetime benefit reduction, the top 40% would see an average of a 4% reduction. And now we know why. (remember the smoothy spiky thing mentioned above?
The Congressional Budget Office estimates that the average Social Security recipient would face an average reduction of $3 per person per month in 2014.
By 2023 that will have grown to a $30 per month reduction in monthly benefits under Chained CPI-U v. CPI-W.
By 2033 outlays for Social Security would be 3% lower than they are now across the board.
The average annual COLA on SS has been around 2.8% over the 50 years or so. Looking back those same 50 years, the use of Chained CPI-U would have resulted in an average of a 2.55% annual COLA increase.
So as you can see, we aren’t talking about a lot of money. Even for the poorest of seniors a $3 benefit reduction doesn’t seem like a death sentence. Overall, the use of Chained CPI-U hardly seems radical. It may even do what is advertised, keep the Social Security trust fund paying out at a 100% rate further into the future.
But remember, this is what happens if Chained CPI-U is applied to Social Security without any modifications or other measures. What about the President’s specific proposal?
The President’s specific proposal is even more gentle:
Understanding that a straight shot of Chained CPI-U will harm many poorer and more elderly Social Security recipients, the President’s proposal contains some provisions that lessen the impact of the changes to the COLA.
The first is a benefit enhancement that would start kicking in at age 76. Social Security recipients would receive a “benefit bump” of 5% of the average Social Security recipient benefit.
Currently that would mean that a 76 year old, or 86 year old or 96 year old would receive about $800 a month in today’s dollars to offset the smaller COLAs under Chained CPI-U
Not only that, but the benefit would be coded so to avoid moving a Social Security recipient near the poverty line into a new classification that could potentially change their eligibility for other federal and state programs.
The average Social Security recipient (the true average) would essentially see a $28,004 dilution of earnings due to more moderate COLAs over a 30 year period.
Under the President’s proposal, that would be more like $16,663 over 30 years.
This amounts to an average of $555.43 per year for 30 years, or $46 per month.
Again, objectively this is not a vast sum of money. If this is a better measurement of inflation, and will help keep Social Security running at full steam longer, then perhaps its worth considering.
(Of course one could then wonder if Chained CPI-U is a more accurate measure of inflation, why would any group need to be protected? Aren’t they simply being given more money than they should? But I digress.)
This change seems to be a relatively moderate, reasonable way to address the longevity of the Social Security program without severely damaging a vulnerable population of benefit recipients. No one wants a long-term benefit cut, but if there are to be changes to benefits, this would be a less painful way to do it.
In addition, the President has wisely tied any change to how the COLA calculation is figured to a broader package of budget proposals. This is not a stand-alone plan or proposal.
So why the outrage?
Social Security has enough money in its trust fund to pay 100% of benefits until around 2033. Even then, the program can keep paying at 75% of that rate until 2086 or so. So obviously something needs to be done and done reasonably soon.
There are plenty of easy fixes to the Social Security funding issue. We could lift the cap on income that is subject to Social Security tax and employ “bends” already in the system to make sure that wealthier beneficiaries aren’t putting money in that they won’t get back from the system. Turning Social Security into a means-tested program would remove its most potent political shield and likely lead to the end or alteration of the program long-term. Here’s how it would work:
- 1. Social Security already employs 3 ‘bend points’ to determine your Primary Insurance Amount at benefit election.
- 2. At 62 each year’s earnings are tallied up & indexed for inflation, & your highest 35 years of earnings are averaged out. ‘Bends’ are then applied.
- 3. The ‘bend’ points determine what your Primary Insurance Amount (SS payment) will be at your full retirement age.
- 4. For someone making an Avg. Monthly Income of $8k over that 35 years, here’s how this breaks down:
- 5. The first $767 of monthly income has a 90% benefit applied to it. The next $3857 of income has a 32% benefit applied. Next $3376 – 15%…
- 6. Or:
*$767 x .90 = $690.30
*3,857 x .32 = $1,234.24
*3,376 x .15 = $506.40
For a total benefit of $2,430.94
- 7. So lifting cap on incomes subject to SS tax would involve simply adding another “bend” to higher incomes.
- 8. Lower income SS recipients already receive a much higher proportional benefit to their incomes. Nothing substantive needs to change.
- 9. Someone making $500k a year for example would still have ALL their income applied to a benefit, but have a “bend” of .001% (for example) applied to their monthly income above a certain amount of monthly income.
Unfortunately, there are no votes in Congress to pass this simple “fix”. Letting the Bush tax cuts on the wealthy expire nearly brought down our government already.
-What about raising the retirement age? Or simply lowering benefits? No votes for that either
-Raising payroll taxes again is certain political death for a big section of Congress for different reasons.
-We could always just vote in a better Congress, but with their approval rating at 13% and a re-election rate of 90%, this is no sure thing and may involve other, even bigger systemic changes to our system of government, let alone Social Security.
Its beginning to look like Chained CPI-U may be on the table simply because its the only thing that can possibly pass.
I don’t think people are really up in arms about an average lifetime monthly benefit cut of $46 a month. Yes it matters for people relying (especially widowed women) solely on Social Security for their income. And it would really suck to lose that extra money.
I think Chained CPI-U may indeed be a more accurate way to calculate inflation. Maybe not for seniors, at least not now, but it could indeed be more accurate.
I think that some compromise may be necessary to get Republicans to agree to other vital changes and enhancements necessary to keep our nation growing and prospering. The GOP and our problems aren’t going to vanish overnight (even if we want them to).
I think The Fuss is coming from a deeper, darker place that is only explained with proper context. In a vacuum, this is not a radical proposal, but this doesn’t exist in a vacuum, it exists in the context of the last 30 years.
After the Reagan tax cuts for the rich, and payroll tax hikes on everyone else
After the removal of Glass-Steagall
After the Bankruptcy Act of 2005
After the Medicare D bill of 2006
After the lowering of taxes on investment income and carried interest for wealthy
After NAFTA and other policies making it easier and cheaper to move jobs and operations overseas than to leave them in the US
After systematic attacks on unions, starting with the grand experiment of firing all of the air-traffic controllers
After the tightening of welfare requirements
Etc, etc etc etc
After all of this I think I may understand the out-sized reaction to a rather modest proposal.
After all of this, it becomes clearer why people are irate about this clerical change to how benefits are calculated.
After all of the naked attacks on the poor, lower-middle class and middle class over the last 30 years, you’ll have to forgive them if any more sacrifices, no matter how gentle are viewed antagonistically.
As they should be.
http://www.fas.org/sgp/crs/misc/RL32293.pdf (great paper on the math and methodology behind Chained CPI-U)