Millions, Billions, Trillions, Who's Counting?
The single change that would most greatly improve economic reporting would be a requirement that all numbers be placed in a context that made them meaningful to readers. We know that many of the numbers that appear in reports on the budget or the economy are not currently meaningful.
For example, seeing that a transportation bill will cost $196 billion over the next six years is absolutely meaningless to everyone but a few budget wonks. If a zero were added or subtracted to this number it would probably mean almost exactly the same thing to most readers. An article could substitute "really big number" for "$196 billion" and still provide as much information. The simple alternatives are to express the sum as a share of total spending or on a per person basis, both of which would provide metrics that are immediately understandable to most readers.
Editorials in the NYT and Post today give us excellent examples of why the current method of expressing these numbers is such bad reporting. The NYT editorial praised congressional Democrats for standing up to President Bush on the war in Iraq: "house Democrats distinguished themselves this week when they stood up to the White House’s latest military funding steamroller: approving only $50 million of the additional $196 million the president requested for the wars in Iraq and Afghanistan."
That's right folks, the editorial said "millions." They meant "billions." Oh well, we all knew that. Of course, it is hugely important that the sum in question is $196 billion and not $196 million. This is approximately 6 percent of projected spending in 2008, about $660 for every person in the country. That is real money -- about 28 times the disputed SCHIP appropriation. If the NYT had expressed the sum in some context that made it understandable to readers, an error of this sort never would have made its way into print. Its editorial could then have informed readers rather than confuse them.
The Post editorial raises the opposite issue, the whole point is confusion. As regular BTP readers know, the Post is engaged in a no holds barred war against Social Security. (I will not comment on the allegations of waterboarding on 15th Street.) The Post warns readers that the Democratic presidential candidates don't have a plan that will fully close the projected $4.75 trillion Social Security shortfall over the next 75 years.
This is intended to sound really scary to readers. After all, as my debating partners on the Social Security privatization circuit always used to say, "that's 'trillion' with a 't.'" Since most people don't have a good idea of what $4.75 trillion over the next 75 years means, it might have been useful to express the sum as a share of projected income over this period. By this measure, the SS trustees projected shortfall would be approximately 0.7 percent of projected income. The shortfall projected by the non-partisan Congressional Budget office is approximately 0.6 percent of projected income over this period. By comparison, annual spending on the wars in Iraq and Afghanistan now comes to 1.3 percent of national income, approximately twice the size of the projected SS shortfall.
In this case, context would directly undermine the goal of the Post editorial. The intent is to scare people, not to provide information. However, if it was standard practice to place big numbers in a context in which they could be understood, the Post would not be able to get away with this cheap trick.
--Dean Baker
Feeds: 


COMMENTS (17)
I've been trying to figure out why the liberal Times and Post are so down and stupid about SS.
The only reason I can think of is that they want more immigration and, mistakenly, believe that it will relieve the SS funding problem. Hence they must emphasize that there is a problem
Posted by: Robert Hume | November 17, 2007 11:16 AM
My theory on their infatuation with Social Security goes back to that op/ed by one of their troupe of morons (Broder perhaps) a few years ago that told some high-school girl that math was overrated and journalists did just fine without knowing how to add or subtract.
This lead to them raising their hands up and bowing down when they hear the REALLY, REALLY, BIG NUMBER: 4.75 trillion! Since they can't compute that over 75 years it is not such a REALLY, REALLY, BIG NUMBER, our Social Security policy is being driven by a deification of big numbers that are over the heads of the WaPo staff.
Posted by: flounder | November 17, 2007 11:33 AM
Read some of John Allen Paulos books on Innumeracy: "Innumeracy: Mathematical Illiteracy and Its Consequences", "A Mathematician Reads the Newspaper"
http://www.math.temple.edu/~paulos/
Posted by: ef | November 17, 2007 12:11 PM
"The single change that would most greatly improve economic reporting would be a requirement that all numbers be placed in a context that made them meaningful to readers."
Keep up the good fight, Dean.
Posted by: F. Frederson | November 17, 2007 12:22 PM
very very nice post. I couldn't agree more.
Posted by: tt | November 18, 2007 3:11 AM
Dean: For those of us who don't carry the budget info around in our head, is there an online calculator that let's us plug in numbers, years etc and will kick out the precentages and dollars per person you helpfully provide in your posts?
Posted by: Martin | November 18, 2007 11:17 AM
One interesting thing about that $4.7 trillion. Although the Trustees call it a 'liability' legally it isn't one at all. It builds in the assumption that the current scheduled benefit is in fact the ideal allocation, it assumes that if and when the Trust Fund goes to depletion that America will be forced to make up the gap.
But after considering the implications of what I call Rosser's Equation (75% of 160% = 120%) this is not clear to me at all. My personal belief is that economic growth going forward is more likely to look like Low Cost than Intermediate Cost, that is we should be able to pay out 100% of benefits for years past 2041. But whenever depletion happens and whatever level benefits have to be cut my current reaction is more like "So what, your real check will still be better in real terms than my Mom's today"
There is nothing magic about the current schedule and it is certainly possible to look and see an 85% or 90% result as being accepteable. The assumption that we would, could, or should simply sacrifice current utility of that payroll dollar for some theoretical future utility is not as obvious as some would like to make it.
Because there are two ways of looking at Trust Fund Depletion. One is the familiar 'Crisis' perspective. But the other is 'Tax Cut'. If as a people we decide that the new level of benefits is in fact equitable then there would be nothing wrong with waving away the remaining theoretical obligation and dramatically reduce the tax burden overnight. Workers should and I think will demand that every penny lent be paid back with interest. After that the claims for social justice get a little muddy. "Sorry 97% is all current income can support" might well be an acceptable answer.
Legally there is no liability.
Posted by: Bruce Webb | November 18, 2007 3:40 PM
Bruce Webb wrote, Although the Trustees call it a 'liability' legally it isn't one at all.
But that's a two-edged sword.
Because current and future beneficiaries have no property interested in their benefits, there's nothing in the way of legality or constitutional right to not be capriciously deprived of property to stop right-wing thugs from canceling/reducing benefits and pocketing payroll tax revenues (to further their program of shifting government taxes onto labor).
Posted by: liberal | November 18, 2007 5:00 PM
Martin,
we have a budget calculator that you may find useful on the CEPR website at http://www.cepr.net/calculators/calc_budget.html
Posted by: Dean Baker | November 18, 2007 6:25 PM
Two things bother me more than the lack of meaning in bare numbers. First, how good is the estimate; what error could be expected at what confidence level? Second, how is spending expected to be distributed over the time period. Frankly, I've seen too many politicians and managers use "X dollars over Y years" to seem to promise something when it's really just a stall.
Joe
Posted by: OJCsr | November 18, 2007 7:31 PM
Neato, thanks.
Posted by: Martin | November 18, 2007 10:36 PM
"Because current and future beneficiaries have no property interested in their benefits, there's nothing in the way of legality or constitutional right to not be capriciously deprived of property"
Liberal at a fundamental level that is true about anything. What power do you have to keep the government from seizing your house at will in the name of the 'public good'? Well you could ask the losing side of Kelo v City of New London on that one. What power as a individual would you have it the President convinced the Fed to simply crank up the printing press and so hyperinflate away your cash assets? In the end there is no asset either 'hard' or 'soft' which cannot be directly or indirectly affected by government action, in the end the only real power the public retains is that derived from the ballot box.
Is Social Security more? or less? exposed to governmental action than anything else. Well in this matter perception becomes reality. If people don't believe that the Trust Fund is real and backed by Full Faith and Credit of the United States, then the door is open for manipulation. If on the other hand people are willing to step up and defend their rights to the dollars in the Trust Fund in the same way they will defend their rights to that piece of paper in their wallet then Social Security is bulletproof.
The mistake is allowing the opposition to redefine what is and what isn't backed by Full Faith and Credit. Because in the end all property rights are a matter of political will by the collective ownership. That house deed in your safety deposit box is worthless if the organs of government are not willing to enforce your rights as needed. Which in turn rests on your ability to change the government when it fails to protect what you perceive as property rights.
Posted by: Bruce Webb | November 19, 2007 11:37 AM
This has become irritating. First the powers what be rattle on about the coming crisis, then many bits and bytes, as well as no small amount of ink, are spent refuting the assumptions. The powers what be go silent. Then they come back again, making the same stupid arguments that people have already refuted. Just this morning I read in Ruben Navarrete's column that one proposal for "saving" Social Security is means-testing.
I will again write to the paper to remind people that means-testing would cost far more money than would be saved by denying the small number of recipients who would be denied benefits. I admire Dean Baker's ability to keep responding. I'm not sure how long I can do it, though.
Posted by: PeonInChief | November 19, 2007 1:17 PM
No problem exists with Social Security. One would need to remove the taxable income cap of 99,000 dollars and increase the payroll tax by a percent or two.
The Republiklan party has never liked Social Security nor Medicare and like a Multi Level marketing scammer they come back generation after generation and attempt to ruin social security.
I have started a new political party called the Liberal Democratic Party of the United States. You can read the web page at http://www.dmocrats.org and you will find that this party works differently from other political parties. Take a look and help enact progressive legislation and end the war where you work as a legislator and you vote on legislation.
Posted by: The Liberal Democratic Party of the United States | November 19, 2007 4:47 PM
Can we use some hard numbers?
The current payroll gap is 1.95%. Raising the cap so that it captures wages at the 90% rate would raise the equivalent of 1.0% of that (per the SS actuaries scoring of the LMS plan). The combination of a 1% increase in payroll tax plus an LMS style increase in the cap closes the whole theoretical gap. Then again any short term growth in Real GDP above 2.6% and above 2.0% longterm closes most of the gap on its own.
Social Security might need a financing tweak at some point in the future, but under current projections nothing at all needs to be done before 2017, mechanically nothing is accomplished by flowing more dollars through a system that is currently in cash surplus.
Posted by: Bruce Webb | November 20, 2007 3:09 PM
One tack that I've used for a while on this subject, which I think I lifted from "Innumeracy", is to ask people to estimate the amount of time taken up by a million vs a billion seconds. Almost no one's close. (The answers are "11.5 days" and "approximately 32 years".)
Posted by: ND | November 20, 2007 9:18 PM
昌盛电池
Posted by: 钢管 | October 27, 2008 6:11 AM