Facebook Charged Biden a Higher Price Than Trump for Campaign Ads(themarkup.org)
themarkup.org
Facebook Charged Biden a Higher Price Than Trump for Campaign Ads
https://themarkup.org/election-2020/2020/10/29/facebook-political-ad-targeting-algorithm-prices-trump-biden
3 comments
Facebook has always done everything it can to make the Alex Jones and Milo Yiannopoulus of the world happy.
Note: the data behind this came from NYU's Ad Observatory project which Facebook is trying to shutdown[0].
[0] https://news.ycombinator.com/item?id=24909056
[0] https://news.ycombinator.com/item?id=24909056
Conclusion from the linked analysis:
> We found Biden, on average, paid more for his ads than Trump did. Facebook says that pricing for ads varies based on parameters set by the advertiser and Facebook’s instant auctions system. The company has also said in the past that it subsidizes ads that its algorithms view as more engaging. Without greater transparency as to how Facebook prices ads, we cannot say conclusively why the difference in prices charged to the presidential campaigns occurred.
I’m trying to connect this with a point made in the article, itself:
> The sort of differential pricing for political advertising that The Markup found would be illegal or unconventional in other media. Federal laws require TV stations to charge candidates the same price—the lowest that they charge any advertiser—for ads. Some states forbid newspaper publishers to charge one candidate a higher price.
But what happens if candidate 1 chose prime time slots and candidate 2 did not...seems unlikely they would be charged the same price. The analysis also says
> Facebook does not disclose in its ad library one key piece of information: whom the ads were targeted at using its audience microtargeting tools. Nor does it disclose a campaign’s “objective”—whether it’s just getting its ads in front of the voters, or other goals like raising money, gathering email addresses, or getting users to watch a video all the way through.
So: “Biden paid more and we don’t have enough info to say why, let alone whether this is somehow unfair.” Is that it?
> We found Biden, on average, paid more for his ads than Trump did. Facebook says that pricing for ads varies based on parameters set by the advertiser and Facebook’s instant auctions system. The company has also said in the past that it subsidizes ads that its algorithms view as more engaging. Without greater transparency as to how Facebook prices ads, we cannot say conclusively why the difference in prices charged to the presidential campaigns occurred.
I’m trying to connect this with a point made in the article, itself:
> The sort of differential pricing for political advertising that The Markup found would be illegal or unconventional in other media. Federal laws require TV stations to charge candidates the same price—the lowest that they charge any advertiser—for ads. Some states forbid newspaper publishers to charge one candidate a higher price.
But what happens if candidate 1 chose prime time slots and candidate 2 did not...seems unlikely they would be charged the same price. The analysis also says
> Facebook does not disclose in its ad library one key piece of information: whom the ads were targeted at using its audience microtargeting tools. Nor does it disclose a campaign’s “objective”—whether it’s just getting its ads in front of the voters, or other goals like raising money, gathering email addresses, or getting users to watch a video all the way through.
So: “Biden paid more and we don’t have enough info to say why, let alone whether this is somehow unfair.” Is that it?
> So: “Biden paid more and we don’t have enough info to say why, let alone whether this is somehow unfair.” Is that it?
That's the subtext, but from TFA this is explicitly illegal in other media forms -- presidential candidates all always pay the lowest ad price.
That's the subtext, but from TFA this is explicitly illegal in other media forms -- presidential candidates all always pay the lowest ad price.
Regardless of slot and placement?
My assumptions would be for a given type they would get the lowest price. e.g. on-peak vs off-peak pricing would be two (2) different plans and ask different prices of which both could be lowest for that given specification.
For TV, based on time-slots, location, etc there are normally different tiers that have a range based on type of ad, length, play count, etc. I would imagine for each of these tiers a candidate would pay the lowest possible for each tiers' bracket.
For TV, based on time-slots, location, etc there are normally different tiers that have a range based on type of ad, length, play count, etc. I would imagine for each of these tiers a candidate would pay the lowest possible for each tiers' bracket.
The point stands that political ads on broadcast stations are regulated and those on social media are not. That is problematic. However, FCC regs seem to allow stations to charge different prices based on the “class” of the placement. Not sure what the right social media equivalent of “class” would be. Seems reasonable that it might include targeting and campaign objectives. Which we don’t know in this case. So seems a stretch to imply that FB is charging a higher price for equivalent ads. They may be, but IMO the authors don’t have sufficient evidence based on what they could access.
> A candidate shall be charged no more per unit than the station charges its most favored commercial advertisers for the same classes and amounts of time for the same periods.
https://www.fcc.gov/media/policy/statutes-and-rules-candidat...
> A candidate shall be charged no more per unit than the station charges its most favored commercial advertisers for the same classes and amounts of time for the same periods.
https://www.fcc.gov/media/policy/statutes-and-rules-candidat...
>So: “Biden paid more and we don’t have enough info to say why, let alone whether this is somehow unfair.” Is that it?
It's basically this. Biden paid more in the past, but at some point in September things crossed and now Trump is paying more.
Audience targeting also seems to have a big impact on ad cost for FB, so the only real conclusion to be drawn is that the ad rates changed, possibly because of different targeting parameters. That feels very similar to comparing costs of televised ads while not knowing what time slots were involved.
That said, one ad marketplace is far more opaque, so it's hard for an outside observer to accurately gauge fairness.
It's basically this. Biden paid more in the past, but at some point in September things crossed and now Trump is paying more.
Audience targeting also seems to have a big impact on ad cost for FB, so the only real conclusion to be drawn is that the ad rates changed, possibly because of different targeting parameters. That feels very similar to comparing costs of televised ads while not knowing what time slots were involved.
That said, one ad marketplace is far more opaque, so it's hard for an outside observer to accurately gauge fairness.