Many have warned about the grave dangers of “money in politics” in the US. However, does this theory fit the data? Let’s test it out.
Since this is a test case, we should make it really really clear-cut. For example, even if money is super-important in politics, any one billionaire won’t get everything he wants. There could be another billionaire wanting the opposite. But if a whole bunch of billionaires got together, and said they all wanted the same thing, then you’d expect them to win if money really mattered.
Likewise, if a billionaire wants something, but he’s quiet and keeps it to himself, nothing will happen. So we should choose a case where billionaires are being really loud and actively lobbying for something. That way, we know that Congress knows what they want.
And similarly, we should pick something that’s at least moderately popular, and isn’t too big of a change. Even if Congress were in the pockets of Charles Koch or George Soros, they still wouldn’t vote for something really extreme, like 99% tax rates or abolishing the military. There are about 200,000 pages in the US Code, so for our test case, let’s also pick a proposal that changes, say, four or five of them.
Fortunately, the last year has given us a perfect ‘experiment’. Nine of the richest companies in the US, with a total valuation of $1.4 trillion, all got together and said exactly what they wanted. To make sure everyone knew, they even took out a full-page ad in the New York Times. And their idea was even pretty popular, with 60% of Americans in favor.
Result: nothing.
Okay, so maybe powerful national politicians can’t be pushed around by wealthy corporations. But let’s take an even more extreme example. Let’s take a big, powerful, multi-national corporation, with piles of money and powerful friends, against a little local government that nobody knows about. And sometimes the media or the courts will “fight for the little guy”, so let’s take a case where they aren’t involved.
Last year, Google wanted to build a bridge over a creek, between two Google office buildings. Nothing special, just an ordinary twenty-foot bridge, to save people from making a long, polluting car trip on congested Highway 101. But, to build it, they needed the permission of the Mountain View City Council. Result: denied.
The denial was ostensibly for environmental reasons, even though the bridge would save thousands of miles of polluting car travel. So Google offered to conduct an environmental impact review, to prove the bridge was green-friendly. But even that was denied.
To illustrate how extreme the disparity is here, let’s examine a few key statistics:
Google budget: $45,860,000,000
Mountain View budget: $87,000,000
Google employees: 49,430
Mountain View employees: 378
Google is also Mountain View’s largest employer, and owns 11% of Mountain View’s real estate by valuation. According to the money-in-politics theory, the City of Mountain View should be bending over backwards to please them. Yet, this isn’t what we see at all.
The money-in-politics theory has been formally tested, by famed economist Steven Levitt. Political campaigns are normally hard to experiment with, since there are many “uncontrolled variables” – did Barack Obama win because of all his donations, or did people donate because he was a popular candidate? And so Levitt looked at were races where the same two candidates ran against each other multiple times. He found that, in Congressional races where candidates spent about $250K (1990 dollars), every $100K spent got another 0.3% of the vote, a tiny amount.
Reading the newspapers, one hears about powerful people making political donations. One might surmise that the donations cause the power – first you donate, then you become powerful. However, looking more closely, this seems like a cause-and-effect error. People, for the most part, first become powerful, through some as-yet-unknown process. Then, after they have power, they start donating to campaigns. Attempts to do it the other way around have tended to not work so well.
The Rationalist Conspiracy 
Your point is a strong one, given the evidence provided in the latter examples, but the first one (the tech companies advocating surveillance reform) is probably a bad example because 1. The tech companies don’t actually have a very strong profit motive for wanting such a reform, making their “request” a pretty transparent public-relations move, and more importantly 2. they were opposed in this matter by the US government’s intelligence agencies, which have at least as much legislative clout in the US as major corporations
On the other hand, there’s this.
I haven’t the time to do a detailed statistical analysis of this, but the conclusion (“Vote Democrat!”) makes me veeeeeeery suspicious. Someone saying, “And therefore, if we just elect Party X, everything would be fine!” should be read in exactly the same tone as a used car salesman saying “And therefore, if you just buy now, everything in your life will be fine!”. See https://rationalconspiracy.com/2013/06/20/politics-is-suspicious/, http://lesswrong.com/lw/gw/politics_is_the_mindkiller/.
> He found that, in Congressional races where candidates spent about $250K (1990 dollars), every $100K spent got another 0.3% of the vote, a tiny amount.
How would one know if 0.3% is a ‘tiny’ amount given that Levitt did not take the additional step and calculate how much 0.3% increases the odds of victory for the challenger? Especially given that in a two-party race, the challenger winning 0.3% means the incumbet loses 0.3% for a shift of 0.6%, and many races being close and down to single %s?
Naively, if $100k buys you 0.3% of the vote, $10M buys you 30% of the vote. An excellent return, if you can afford the initial investment.
I’m pretty sure that if Google’s owners decided to spend $100M on campaign contributions, they could completely replace Mountain View’s entire city council with their own preferred candidates.
How much money did Google (or its owners) actually spend on influencing the composition of the Mountain View City Council over the last 10 years? That would be a far more relevant figure than the amount of money that the company is worth.
I don’t know how much they spent, but if they didn’t spend anything, that leaves the obvious question of “why not?”. Google is willing to spend billions on patents, billions on employee perks, and billions on physically building their offices (their balance sheet shows $20 billion in “Property, Plant and Equipment”); why would they not spend tiny amounts on lobbying, if that would be an effective way of solving their problems?
Before the Microsoft trial, Microsoft didn’t spend anything on lobbying. That’s why the trial managed to make it to court in the first place–Microsoft just had a blind spot about how beneficial lobbying would be. In other words, they weren’t rational. Who’s to say Google isn’t rational either?
Do you have cites for that?
Here is a typical account, from 1999.
The timeline is a bit off. MS dipped its toe in lobbying in 1995. The spending jumped during the problems with IE, but that wasn’t enough to head off a trial. Whether earlier spending would have made a difference is hard to judge, but the lobbying spending remained high, suggesting that the company regretted its earlier low spending.
MS had an anti-trust settlement in 1994. It hired Jack Krumholtz as its first lobbyist in 1995; he seems to agree with the basic claims. I don’t know if they hired outside lobbyists before that. He was not initially working on anti-trust, so perhaps the 1994/1995 timing is a coincidence.
Not to mention Google doesn’t want to entirely piss off the city in which it resides. Its not in their best interest.