Don’t overestimate spending on elections

Binyamin Appelbaum has a new and excellent piece on this topic:

Even the 2012 presidential election, which recorded $2.6 billion in campaign spending, underperformed many forecasts. And spending has declined in each of the last two congressional elections. Candidates and other interested parties spent $3.7 billion on this year’s midterms, down from an inflation-adjusted total of $3.8 billion in 2012, which was less than the $4 billion spent in2010, according to the nonprofit Center for Responsive Politics. (These figures do not include a few hundred million dollars in unreported spending on issue ads.) In fact, spending has dropped as the economy has grown and despite a series of contests in which at least one house of Congress was plausibly at stake. “Dire warnings rang out that the decision would herald a new era in politics,” wrote Adam Bonica, a Stanford University political scientist, in a 2013 paper about the effects of Citizens United. “Three years on, there is little evidence that these predictions have come to pass.” Over the past year, Americans spent more on almonds than on selecting their representatives in Congress.

The article is here, interesting throughout.  Campaign finance, of course, is one of the areas where “the Left” is most likely to take an anti-science stance.


Niall Ferguson (once again) has an excellent Chap. 8 in his 2001 book "Cash Nexus" on Elections and economics. It's too dense to summarize, but here is one choice factoid: "Total expenditure by candidates on the 1880 election exceeded 1.7 million pounds. In 1997 prices this amounts to over 20 pounds per vote, compared with a figure in 1997 of 42 pence. In relation to GDP expenditure by candidates on general elections has fallen by an astonishing 98% since 1880 (Fig. 22)"

"Over the past year, Americans spent more on almonds than on selecting their representatives in Congress. " - I think this is a teeny bit (teeny, I like that word) dishonest because I read there's an almond shortage at the moment due to the California drought, that's pushed up prices.

That only makes it all the more impressive, since the amount spent on almonds equals the price times the quantity supplied.

True, all though bear in mind that in 1880 Britain was the most powerful country in the world, and only a sixth of its adult population could vote; bribing voters was not yet illegal (1883), and the secret ballot had only just been introduced and was not well enforced. You got a lot more bang for your buck in the old days.

100% correct. I spent exactly $0 on electing my idiot representative (I repeat myself again).

Hilarious! Because elected representatives are so damn stupid! LOLOLOLOLOL....

because in China the CCP cadre school produces reliably elite technocrats.

Do they link to the data source for this? 2012 seems quite a bit off base given there were many competitive races. Also I think while one prediction would be that more and more money would be spent on election I think there were lots of issues "the left" has with this form of financing. How is spending by 501(4)(c)'s tracked?

Also now that individuals are able to spend more does that increase their influence? It's not just a raw numbers game. I think most on "the left" would feel better about everybody donating $20 to some political campaign if that resulted in 3 billion dollars in total spending than the total spending being 3 billion dollars where one guy spent 2 billion dollars and encouraged those running to jockey for his support heavily before he declared how he would spend. Do you know how many copies of his or her own book (to send to donors of course) a candidate can buy with 2 billion dollars?

Why don't you try to influence a local race and local politicians right up against the boundary of the law to see if it's possible? Which outcome would surprise you more, by how much?

I don't think it includes 501c4 dark money either. See below.

IT'S SCIENCE YOU DENIER! I'll bet your skepticism is motivated by racism and rape denial.

So, three years of 'evidence' can be extrapolated? for how far in the future? This science? Trash!

Also, "We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run." Further - We tend to be stupid in the medium run.

"Don’t overestimate {effects of} spending on elections"

How did people get elected to office before TV existed ?

Since most campaign spending is for TV, is TV actually so critical to getting elected ? What's the objective evidence overall ?

Elections are cheaper than they ought to be. A collective action problem? This is one reason why in countries with massive trade union confederations (e.g. Scandinavia) Social Democratic parties can do so well. Half the population effectively pays a party tax.

How does this square with papers I've seen showing that the ROI on lobbying tends to be 100% or more? If lobbying has such high returns, why do so few people spend on it? It seems like the plausible explanations must be some combination of: A) Due to personal ethical and moral qualms businesses aren't lobbying even if it would be in their self-interest. B) The previous research overstates the returns to lobbying, or the marginal lobbyist sees much lower return than the average lobbyist. C) Businesses are simply unaware of potential lobbying opportunities.

There was an article on this in The Public Interest about a dozen years ago. The thesis was that lobbying efforts are toward the end of preserving extant privileges, not seeking new privileges, so the population of lobbying concerns will be drawn from today's clientele (real estate, agribusiness, higher education, the medical sector, intellectual property traders &c).

Ahhh, that certainly seems to make a lot of sense. Thanks for the reference. I suppose that brings up the question of why its so much politically easier to preserve existing rent-seeking privileges compared to creating new ones? Status quo bias on the part of voters?

I think their point was that politicians generate the privileges for the various reasons they have, then trade associations form to protect them.

Lobbying is a very different activity than electioneering.

Also, any effective lobbyist must limit his activities to the range of policies that are politically achievable. This significantly circumscribes the scope of lobbying far more than many "clean governance" activists would be able to admit.

Campaign finance, of course, is one of the areas where “the Left” is most likely to take an anti-science stance.

No, they're more likely to hew to transparently phony standards: spending by our guys is good and spending by the opposition is illegitimate (see the repetitive talking points re the Koch brothers, whose campaign spending is dwarfed by dozens of other parties).

Not sure what the difference is in methodology, but I am not seeing any decline:

Are those numbers adjusted for inflation? If not, then when adjusting for that, 2012 was less expensive than 2010.

2012: $3,664,141,430,

2010: $3,631,712,836 (cited), $3,823,895,241.42 (adjusted by

Inflation adjustment.

2012 had a national presidential race, though. The last comparable Congressional race to 2010 was 2006. I would also say that races involving a President running for re-election are more comparable than an election where there are two new nominees running. I also think that the gerrymandering of House districts has more of an effect on fundraising than any effect of inflation or campaign finance regime (or lack thereof).

Umm, gerrymandering is not new, so I'm hard pressed to think how it would uniquely affect recent races. Also, the original article compared two off-cycle elections.

Considering all the hyperventilation over Citizen's United, this is pretty damning empirical evidence, as the quoted article points out.

You don't see a trend of a decreasing number of contested/competitive House seats, where the political considerations involved in the creation of district boundaries are/were a factor? You don't think that a decreasing number of contested/competitive House races would have a chilling effect on fundraising for House races, especially when compared to inflation or campaign finance reform, especially over a period of only a couple of election cycles?

A trend that correlates with the money spent? No. The gerrymandering and "a trend of a decreasing number of contested/competitive House seats" has been going on for at least thirty years in earnest, and probably much longer. I remember being taught about gerrymandering and seeing a map of a congressional district back in, maybe, middle-school. Meanwhile, the cost of elections goes up, then seems to flatline and maybe even go down. Zero correlation, whatsoever.

Gerrymandering is real, and it may even be bad, but that doesn't make it the root cause of the crusade du jour.

No, I don't. Re-election rates a generation ago were higher than they are today. In New York (when I was involved in local politics), those for the state legislature were over 99%. That was thirty years ago. It cannot get any worse than that.

The tendency toward gerrymandering was exacerbated by court decisions which require max-black district maps and require absurdly small differences in district populations.

For all the backing and filling, It's having less of an effect than it once did. In 1984, the political parties split the congressional vote almost 50-50, but the Democrats won 58% of the seats. Of course, you never heard complaints about gerrymandering from partisan Democrats at that time.

I never wrote "root cause". I intimated that the construction of House district boundaries for political reasons was a greater factor in the lack of contested-ness in House elections than any very recent change in campaign finance regime (one way or the other). I also don't see how one New York Times article based on a study which admittedly fails to take into account any possible increase in spending on issue advertising is in any way "pretty damning empirical evidence" for anything.

I guess Chris Hughes is just being pissy that his hubby lost.

"(These figures do not include a few hundred million dollars in unreported spending on issue ads.)" How does one know if it was "a few hundred million" or a few billion if it's unreported?

That spending can be monitored with a teevee.

How is the anti-spending left more anti-science than those who spend? They are both making the same assumption: that spending influences outcomes.

I'm fairly okay with the idea that both people who hyperventilate over the need to impose political speech restrictions and those who dump tons of moneys into campaigns are fools.

But we could also suggest a marginal argument. I don't think that anyone doubts that it is expensive to mount a successful major campaign. Clearly you can not become a Senator for $100 or $1,000 no matter what your creds. But if we believe that there is a sharp decline in the marginal value of electioneering spending past some threshold, then we might believe that it is rational to spend a certain amount of money, but irrational to try to restrict further moneys into the system.

Clearly you can not become a Senator for $100 or $1,000 no matter what your creds.

William Proxmire used to spend $350 a pop on his re-election campaigns. He said most members of Congress could do this but elect not to.

I said "become" a Senator advisedly.

Where's the link to the piece.

I don't think the reporting period for contributions is closed sufficiently for anyone to make these conclusions.

Dark Money not included

Search for Appelbaums article and found:

"As the Internal Revenue Service contemplates new rules to illuminate “dark money” in politics, a little-known nonprofit group is fighting back using money traceable to billionaire brothers Charles and David Koch, drug makers and the cable television industry.

The group, American Commitment, received 87 percent of its $13 million in funding between 2011 and 2013 from three Koch-connected nonprofits: the Freedom Partners Chamber of Commerce, the Center to Protect Patient Rights and Free Enterprise America, according to tax filings reviewed by the Center for Public Integrity."

Here is the link:

If you don't include DARK MONEY, you haven't honestly discussed the subject.

Re Dark Money:
"As a “social welfare” nonprofit organized under Sec. 501(c)(4) of the tax code, American Commitment is not required to publicly disclose its donors, even though it overtly supports political candidates."

Blah blah blah Koch brothers blah blah blah. That amount of money will finance two congressional campaigns.

OK, Art, since you know, please tell me how much money the Koch brothers spent last year on elections and political activities and political 501c4 funds.

Answer below.

I enjoyed this piece, but left it wondering if the problem isn't the amount of money in politics, but the willingness of politicians to be influenced cheaply.

An 'old reliable' @ MR:

Cowen as Professor Pangloss.

Bonus Points: with a gratuitous swipe at the 'left': 'campaign finance, of course, is one of the areas where “the Left” is most likely to take an anti-science stance.'

- See more, including references to the Koch Brothers, at:

(These figures do not include a few hundred million dollars in unreported spending on issue ads.) Not to preemptively take a lefty "anti-science" position, but couldn't that make all the difference if that is where there has been growth? What share of the relevant spending is really captured by these figures? Is anything else being left out? Just taking a scientific approach.

Issue ads are not always relevant to an election (granted most of these ads in NC wanted us tell Kay Hagan our thoughts on 'Obamacare' or 'ISIS'. Environmental interests ran issue ads against NC governor Pat McCrory regarding coal ash ponds. He is up for election in 2 years.

Yes, but I thinI the issue ads they refer to are the ones that were disclosed. Issue ads that are not within 60 days of an election do not have to be reported, which is part the reason you find them at other times. Of course the campaign never really ends.

In an auction, I simply need to offer more than my competitors. Offering a lot more is silly. You need to appraise the resources available to the bidders before assuming they all have a real chance at buying the item. That's Lobbying. On the other hand, in a political campaign, you must look at the resources available to the candidates, and how the campaign is seen to be progressing, through polls, of course. I might intend to spend $10,000 to defeat my opponent, but, if I find myself falling behind, I might choose to spend $6000. My point is that it's relative to the campaign.

If the point is that people should be spending a lot more on lobbying and campaigning, you also need to remember that you simply need to spend more than the other donors. Then there's graft, which is apparently hard to distinguish from forgetting to post the donation or whatever. And sex. It's a much more messy problem than appears at first sight.

I should probably admit that I would prefer being bought off by sex, but I'd rather not commit myself until the opportunity actually arises.

Should this be filed under "average is over"? We're also seeing a lot of record-smashing (not just record-breaking) amounts spent on particular elections.

A lot of touchiness by people who don't want to accept scientific facts.

If the data is wrong because of dark money, prove it.

Well isn't that the thing about dark money? What methodology do you propose?

A methodology that gets data good enough to justify laws that restrict speech, at the very least.

Independent Spending continues to hit records.

Comments for this post are closed