All 4 books by Edward Tufte now in
paperback editions, $100 for all 4
Visual Display of Quantitative Information
Envisioning Information
Visual Explanations
Beautiful Evidence
Paper/printing = original clothbound books.
Only available through ET's Graphics Press:
catalog + shopping cart
All 4 clothbound books, autographed by the author $150
catalog + shopping cart
Edward Tufte e-books
Immediate download to any computer:
Visual and Statistical Thinking $2
The Cognitive Style of Powerpoint $2
Seeing Around + Feynman Diagrams $2
Data Analysis for Politics and Policy $2
catalog + shopping cart
Edward Tufte one-day course,
Presenting Data and Information
Boston, March 14, 15, 16
Oakland, April 20
San Jose, April 21
Palo Alto, April 24
San Francisco, April 25, 26
How much is $87 billion for Iraq? Update: How about $1.2 trillion?

Anyone care to comment? I think there is always some bias in the choice of comparisons in this kind of graphic, but this one seems relatively clean.

-- Alex Drahon (email)

How much is $87 billion?

Compare the Washington Post graphic with a similar one in the 9/9/03 New York Times:

While the data presented is admittedly not entirely 1-to-1, the Times graphic conveys much of the same idea with easily understood percentage comparisons, and much less chartjunk (note especially the Post's "Bush War on Terror Proposal" on a black banner). The Times also makes comparisons to other wars/conflicts with inflation-adjusted figures, and provides some break-down of the $87 billion.

The Times' version seems to present more and better data for readers to examine, regardless of their political views.

-- Mark Kasinskas (email)

How much is $87 billion?

Back in the days when I was writing about and teaching political economy (remember that classic book, Political Control of the Economy?), I figured out a rule of thumb for teaching my students how to get hold of the meaning of X billion dollars in government spending. It is approximately X dollars per month per U.S. household for a year. So each household, on average, would contribute $87 per month or a $1000 a year for Iraq (ignoring deficit financing costs and data flimflam).

As I recall, this works reasonably for only a few more years; it was fairly accurate in 1980. Could someone please check out the data and the arithmetic to be sure this makes sense?

-- Edward Tufte

How much is $87 billion?

The Times does a more thorough job in breaking down where the money is going, though its presentation of the numerical quantities could have been done in a more tabular, accounting form, with component amounts lined up with the total amounts. The eye goes right, then left, then right again to follow the numbers. Not optimal. Instead, put'em in a column, with subtotal items slightly offset.

In addition, the Times does a better job putting this $87 billion in the context of the overall Federal Budget -- $87 billion seems like a lot, but at 4%, it's a sliver. Juxtaposing it with the inflation-adjusted costs of other famous/infamous wars/programs is also enlightenting, though here they might have informed which inflation-adjusting compounder they used (CPI?).

My bias in presenting numbers is, if possible, show calculations; if that takes up too much room, be explicit in how you did calculate it so others can replicate the calculations if they so choose.

-- Karl Keller (email)

Response to How much is $87 billion?

The debt clock site put the US population at 292 million people.

I did a little searching at the US census site to determine that the average household size is 2.5 people (average family size is 3.1) in Massachusetts. I couldn't figure out how to check the national average, or total population. That's 117 million households.

So, a billion dollars is $3.42/person, or $8.55/household. $87 billion is $743/household.

There might be a different (and more meaningful) way to calculate the impact on the average household. I would imagine looking at how the current tax burden is spread across households by income tax bracket, then prorating a billion against that, and reporting the amount paid by the average (mean) household. I'll leave that as an exercise for the reader. ;)

-- Mike Combs (email)

Response to How much is $87 billion?

I think most of the facts for the above analysis can be found at The Tax Policy Center.

It looked to me like the mean taxpayer bracket (45%) is 15%. That corresponds to $26K in income for single, $52K for married. $26K is in the second quintile, $52K is in the third quintile. The second quintile pays 3.9% of the total federal tax burden, the third pays 10.2% of the total burden.

So, I think you'd adjust the numbers by these percentages, making the average cost of $1 billion in gov spending only 3.9% of $3.42 for most individuals. That's $0.13/year.

Which makes me strongly suspect my math or methodology. Perhaps someone can correct me.

-- Mike Combs (email)

Response to How much is $87 billion?

The answer to this, and all other economic/tax related questions, is .... it depends. Stating $87 billion as an amount per person/taxpayer/household does get the amount in more digestible terms but doesn't address who is paying how much of the cost since income taxes are paid on a graduating rate scale. For instance the highest quintile of family economic income in 2000 paid 65% of total Federal taxes. This quintile of income is earned by a relatively small number of families. For instance, in 1999 8% of the tax returns filed generated 62% of the tax. Then there is the matter of deficit spending, or spending more than the amount of tax collected. That cost generally gets shifted into the future to be paid by a different collective of taxpayers.

Of the two newspaper presentations the NY Times does relate it to approximated, inflation adjusted, costs of other wars and post-conflict efforts. While that doesn't state how much my personal share is, it does get it into some type of context. The 4% of the entire proposed budget is also a relevant comparison, much better than comparing to other selected sectors of the budget.

Finally, a significant portion of the $87 billion is military related costs. These costs would not likely be zero if there were not a war.

Just a few things the answer depends upon.

-- Gene Prescott (email)

Response to How much is $87 billion?

For honesty's sake, shouldn't these data include the estimated $60B already spent in the "War on Terror"?

Just asking.

-- Pierre Scalise (email)

Response to How much is $87 billion?

I think the comparison to other programs (education, jobs, unemployment, homeland security, environment, healthcare, environment, etc.) is probably the most relevent, based on cumulative costs. But, since I started this trying to figure out the cost to the average taxpayer, I'm going to make one last post in that direction.

It's estimated that 100 million personal (not corporate) returns were filed in 2003. That makes the cost of $1 billion about $10/year per tax filer. Data for this comes from the same tax data website I listed above.

The more accurate numer is to figure it by the burden to the average (mean) taxpayer, based on the income quintiles discussed above. That would put the burden of $1 billion on taxpayers filing as single, which have a mean income of $26K, at $0.40 per year. The burden on joint filers, which have a mean income of $52K, is $1.02 per year.

Thus, the bulk of the costs are paid by higher income taxpayers. But the cost to a typical family of the cumulative cost of the war ($147 billion) is $147. Their typical tax burden is $13,520, so this is an increased burden of 1.1%.

-- Mike Combs (email)

Response to How much is $87 billion?

It seems to me that the discussion of how to put $87 billion in perspective gets away from the main political point, which is whether $87 billion would be better spent on education, roads & bridges, domestic security, air traffic control, health care, or any of dozens of other priorities. It's not so much the amount that matters, it's the relative priority.

The decision to raise revenue by levying taxes occurs independently of decisions which spend that revenue. Regardless of how (or whether!) that $87 billion is spent, we're all already paying for it.

-- Scott Zetlan (email)

Response to How much is $87 billion?

$87 billion expressed in terms of domestic policies (item 3 on the linked page):

-- Edward Tufte

Response to How much is $87 billion?

Though I did the math on the per person costs, I don't think that's a very revealing way to portray large government expenditures. On that scale, there's no way to make the average taxpayer care about any expenditure under about $100 billion, because it costs less than $4/year. This obscures the alternatives available for $4/year or less. "Hey, for less than the cost of a ham sandwich, we can invade Iraq?" Vs. "Hey, by invading Iraq, we have to give up unemployment benefits AND universal healthcare."

I think we've rounded up a good list here of alternative ways to put these numbers into perspective, and I want to recap them and add a few more. I think this kind of information for comparison is probably at the heart of any good executive reporting system.

Avg per person
One problem is that the "average" person is heavily weighted by top taxpayers.
Avg per avg (mean) person
Disguises the cost of the program in terms of opportunities and alternatives.
Costs of alternative programs
What would the cost of continued inspections and military no-fly zone patrols have been?
Costs of other programs
Comparisons to healthcare, environment, unemployment, defense modernization, homeland security, etc.
Comparisons to other wars, occupation costs.
Portion of budget
Comparison to total budget, total discretionary budget, etc.
Other budgets
Comparison to state budgets, other national budgets, etc.
Time to pay off
Including finance charges, how long will it take to pay off this debt? This is one I always look at when I make a big purchase on credit.
Previous estimates
Was the war self-financing? Who made estimates, and how much were they? Did the estimates get more accurate as we got closer to the war? This can help determine a confidence factor for the current projections. I just finished paying a contractor for building a new bathroom in my home, and you can bet I was interested in the actual versus estimated price.

Does anyone have any other things expenses could be compared to? I'm always gathering data like this to enhance the design and data content on executive reports.

-- Mike Combs (email)

Response to How much is $87 billion?

An actual discussion about priorities and opportunity costs in Congress!

-- Edward Tufte

Response to How much is $87 billion?

It will add up to a lot more than $87 billion. My colleague William Nordhaus at Yale made various estimates of the cost of the war (before the war started). Links are at "The Economic Consequences of the War with Iraq":

Note the effort to attribute various costs to specifics and to find ranges of possible costs.

-- Edward Tufte

Response to How much is $87 billion?

Why is it that readers get quantitatively stupid and need Real Big Type in tables that are in the international news section? Compare these tables from The New York Times, September 9-12, 2003; all reproduced at same size. Shown are the Iraq budget table, routine financial data, and sports tables.

I think the explanation rests with the reporters and editors, not the readers. Sports and financial people can count; international reporters and editors don't deal with quantitative evidence.

The median amount of data in a PowerPoint table is 12 numbers (based on 28 PP textbooks). And the PP guidelines at the Harvard School of Public Health recommend 12 numbers per table. Thus these 3 tables can be converted into PowerPoint slide equivalents:

Iraq costs table: 3 PP slides
Baseball data, 2 teams: 117 PP slides
World Stock Funds: 11 PP slides

-- Edward Tufte

Response to How much is $87 billion?

I found an estimate of $288 Billion (in 1940s dollars) for the cost of WWII ( Adjusted for inflation, this is about $2770.55 Billion (

So, the $87 Billion dollar budget for the present program in Iraq is about 3.1% of the cost of WWII.

This seems like a sensible comparison to me. The scope and meaning of WWII is well-known to most people and, more, it is "real" to them the way a can of tuna or a year's income is real. At least for me, this comparison works where comparisons to other government budgets (too big and vague) or to everyday expenses (too small and numerous) fail.

-- JS

Response to How much is $87 billion?

Regarding Sports vs. Financial vs. "News" numbers:

I don't think the relative intelligence of the editors is really the issue. That would infer that sportswriters are more mathematically savvy than the finance department which hardly seems likely.

I think the answer is more likely to lie with the intent of the presentations and the repetition of the sports and finance figures. The "news" presentation is a one-off. It has to be self explanatory and it needs to attract its own readership. It has to compete with the crossword and comics and all of the other news items.

The sports and financial information are repeated day after day. They have a constituency and they have a traditional format which acts as metadata for those who are regular consumers. Old guys with poor eyesight will use a magnifying glass to check the box scores or the daily racing tables. They already know what they want is in there, and they are willing to dig it out.

Also, the sports and financial information are smaller because of the gross quantity that needs to be published.

-- Chris Howard (email)

Our discussion back in 2003 about the costs of the Iraq war--figuring out what a mere $87 billion meant--now seems naive. So, now, what's $1.2 trillion worth?

David Leonhardt, my former student, brings us up to date in his New York Times Economix column: link

-- Edward Tufte

How about trillions?

Annual update of dollar cost of Iraq war here. David M. Herszenhorn, "Estimates of War Cost Were Not Close to Ballpark," New York Times, March 19, 2008.

-- Edward Tufte

Threads relevant to design case studies: