The Inefficiency of Campaigns

Third of a series

If We Weren’t Doing This Now, Would We Start? 

In part one of this series, we considered some of the key issues of vote fraud, and vote counting, in the U.S.  In part two, we looked at the resulting problem of the legitimacy of American elections.  Now, in part three, let’s look at some of the challenges facing candidates and campaigns, starting with the inefficiency of the current campaign system.    

To be sure, at the end of every election, somebody wins, so the system might seem to be at least somewhat efficient.  Yet still, some spectacular examples of inefficiency leap to mind.  For instance, in his 2020 bid for the Democratic presidential nomination, Michael Bloomberg spent some $1.25 billion, all in, and got of course, bupkis.  And in 2022, Citadel’s Ken Griffin donated $50 million to an Illinois Republican gubernatorial hopeful, Richard Irvin, who did all the usual candidate-y things, such as making speeches and running TV ads, and yet he lost badly, coming in third in the GOP primary, garnering a meager 119,000 votes.  Which means that Griffin’s $50 million “investment” worked out to $420 a vote.  

And of course, the real test of the current campaign system will come when something better comes along—as, for example, described here.  Only with competition and the resulting gains in perspective do we really see what works best—and what works worst.  

We can pause to wonder as to what’s going on.  After all, these and other fatcats in and around politics have typically proven that they know how to run operations.  And yet even so, they make spectacularly bad political investments.  Could it be that they are being outfoxed by the campaign consultants?  Or is it that the system just doesn’t work well?   Either way, the industry seems ripe for reinvention and/or disruption. 

When confronted with an issue of the status quo, we can ask the piercing question which Peter Drucker, writing seven decades ago in The Effective Executive, suggested every firm ask of itself: “If we did not already do this, would we go into it now?” And if the answer is “no,” Drucker continued, then the company should start thinking if maybe it, whatever it is, might not be such a good idea.  Indeed, perhaps exit the activity altogether and seek something better.  

According to Open Secrets, total campaign spending, nationwide in the U.S. in 2020, totaled $14.4 billion. Although surely, the amount spent to influence voting is vastly greater than that.  How would, for example, one count all the money spent, directly and indirectly, on political persuasion, be it via organizations or by the media—including social media?  It’s easy to conclude that the true number, if it could ever quantified, is a vast multiple of $14.4 billion.   So there’s a lot of money in politics.  Yet whatever the total, the only thing we know for sure is that the leverage of politics—political influence on the economy—extends to every penny of the $26 trillion or so of U.S. GDP.   So it’s worth figuring out how to do this right.

Craft Businesses

Yet despite all the money and tech—well-funded digital strategists and data-crunchers abound now in political campaigns as budgets for a mere House race swell into the tens of millions—the business itself is still a “craft” business.  That is, there are probably hundreds of thousands Americans who call themselves “campaign consultants,” or professional political operatives of one stripe or another.  They aim to serve the estimated 519,000 elected officials in the U.S.  And yet even the largest campaign campaign operations have but a small market share.  

We can note the enormous churn of creative destruction in the process; campaigns and consultancies come and go with rapidity.   As a result, theres’s a lot of friction in the basic meat of campaigns which is, increasingly, data: the names and addresses of voters and donors.  The national Republican and Democratic parties have their treasured lists, as do the campaign arms of U.S. Senate and House members in both parties: The NRSC, DSCC, NRCC, and DCCC.   And there are similar operations for governors, state attorneys general, state legislatures, and so on. 

In addition, thousands of firms and political action committees have their own proprietary lists.  So we’re already up to thousands of data siloes.  To which we can add the uncounted number of campaigns that pull together at least some data, and do at least something with it, win or lose.  These entities come and go, and their data are often lost, like tears in rain.  We can quickly see that data quality, data security—even basic data retention—is an important issue.  

Truly effective data operations—from Google to the credit card companies to the Social Security Administration—don’t operate this way.  They, and their industries, long ago realized that consolidation into the data equivalent of fortresses was the only way to go, data-sanctity-wise.  In a world of hackers and other kinds of fraudsters, there’s simply no way to avoid a heavy investment in cyber-security. And so those operations that aren’t investing heavily are likely porous.  The wide-open pores are, in fact, legion. 

We can’t say that the status quo is catastrophic; we can merely say that it is suboptimal.  And the lack of privacy and security for personal data is probably destined to collide with some pretty heavy and costly lawsuits and other legal interventions.  This likely future litigation will probably force a roll up—the concentration of the industry in fewer, more defensible forms.  Such concentration is also in keeping with the general tendency of firms to scale up their data operations into larger realms that better capture economies of scale.  

The Roll Up  

To cite a past industry analogy, we might claim that the state of the politics biz today to that of the car industry 100 or more years ago.  Amidst the Cambrian Explosion of auto innovation at the beginning of the 20th century, it’s been estimated that some 2,000 U.S. car companies came into existence in the first two decades of the last century.  And yet by mid-century those U.S. companies were rolled up (willingly or unwillingly) into the Big Three, as well as just a handful of tiny-niche survivors.  Today, there are more big companies than just three, including foreign makers, and yet the top ten brands control more than 90 percent of the market.  

We can see this phenomenon elsewhere. Just a few decades ago, there were thousands of internet service providers, each striving to serve some local market.  Today, there are still thousands, and yet just a handful of “phone” and/or “cable” companies dominate the ISP market.  

It would seem that there’s an inevitability to this process of consolidation, and so it’s likely to come to politics.  In a way, of course, it already has. The two parties, Republican and Democratic, control, between them, 99 percent or more of all partisan political offices. 

Boutique Politics   

The duopoly of politics itself makes the cacophony of political operations all the more interesting—and all the more looking like a lagging indicator.  For the most part, campaign firms style themselves on the persona of the person, or persons, who originate the firm.  Back in 1934, California politicos Clem Whitaker and Leone Baxter started Campaign, Inc.; they are remembered as the first campaign consultants.  

The history of campaign consultancy since abounds with famous people, from Tony Schwartz to Stu Spencer to David Garth to Roger Ailes to Lee Atwater to Bob Squier to Bob Shrum to James Carville to Mandy Grunwald to Jeff Roe.  The commonality of all these trailblazers is that while they and their firms can be hugely profitable in their moment, the companies tend to dissolve quickly.  It’s not that they go bankrupt or cease to exist, it’s that their partners have a way of cashing out or otherwise leaving the scene, while the entities are bought up and and reshuffled. To be sure, sometimes these campaign firms are bought up and rolled up into giant firms.  For instance, in 2010, WPP bought Blue State Digital, bringing the Democratic consultancy into a global multination boasting revenues of $17 billion. There have been hundreds of huge deals, because a hot shop is, indeed, profitable. 

Yet still, there’s the challenge of what the German sociologist Max Weber’s called “the rationalization of charisma.”  That is, the process by which an institution outlasts its founder, if it does.  The charismatic hero will not last forever, and so the company must rationalize his or her charisma, turning it into a system or, if you will, a bureaucracy. It’s the only way something can survive; although, of course, with AI, who knows what new longevity tool will emerge.

Yet we can further add that if a company if depersonalized, it becomes all the more clear that underlying product—the ultimate real currency of campaigning— is data.  That is, data, as expressed in big data, structured in algorithms and interface with CRM, and so on, wins.  

To be sure, there will always be great TV ads—moving, funny, memorable, effective—and there will be internet cognate of this creativity, in the form of videos, animation, AI-ification, gamification, who knows.  But in the end, politics is a numbers game.

Thus the issue of data optimization, even survival, once again rears its head.  With every campaign or corporate move, there’s friction. And that’s trouble, including potential legal liability for data gone wrong. 

The point here is not to argue that a few giant search engines, or phone companies, should be running campaigns. Quite possibly, it’s important, even vital, that campaign creativity and political freedom necessitates continued decentralization.  And yet even if the industry remains decentralized, it will still likely benefit from some sort of consortium element on data, some common repository, the cloud equivalent of a stock exchange, or farmers’ coops, or deposit-insured banks, or reinsurance.   If not a roll up, then a rationalization is coming.

The Fundament Unit is the Vote

And as we think about what might come next, we can learn from the data-driven transformation of another field, which is also a numbers game: baseball.  Perhaps more than any other sport, baseball has always been nerds; if one sits in the stands, one might see someone barely watching the game, but rather, poring over stats, for purposes of betting or perhaps just geekery. 

Seventy years ago, the statistician for the Brooklyn Dodgers, Allan Roth, had a simple-but-profound insight: The fundamental unit in baseball is the run.  That is, stats such as batting average, earned run average, etc. all sounded good—and fans loved them--and yet they weren’t necessarily translating into runs, and thus wins.  And winning is the point of the game, at least to managers and their teams.  What matters is on-base percentage, more than homers and stolen bases and then, of course, a million more stats, such as hits and homers when men are on base.  

There was nothing complicated about Roth’s basic insight, and yet it was profound; even as the complexity  Indeed, the application of his insight mostly consisted of chopping away extraneous statistics—what a scientist would call epiphenomena—focusing on those stats that were actually causally linked to whether or not the team won.  Isn’t that so often the case: One must reduce clutter to actually see what needs to be seen.  In more recent decades, Bill James turned Roth’s simplicity into a new kind of complexity, as he started developing statistics that more closely correlated with runs, such as whether or not a batter hits with men on base.  And then, Billy Beane synthesized Roth and James and created a new style of baseball managing/general managing.  The rules didn’t change, at least not much (the designated hitter being one big change).  But what did change was the evaluation of the data.  (Beane was made famous, of course, in Michael Lewis’ book, Moneyball, and then further immortalized by Brad Pitt in the 2011 Hollywood movie.) 

Ockham’s Election

What Roth did was wield Ockham’s Razor.  Back in the 13th century, William of Ockham said famously, “eliminate unnecessary complexity.”   As with Roth and the run, Ockham’s statement was profound its simplicity.  Ever since, Ockham’s Razor—everything else being equal, choose the simplest possibility—has been the standard for science and logic.  (And as such, it was loathed by others of a different turn of mind; the conservative Richard Waver, author of the 1947 book, Ideas Have Consequences, argued that Ockham was when the world took a decisive  turn for the worse because the Razor began chipping way at faith.) 

In subsequent centuries, figures such as Sir Francis Bacon further clarified the scientific method, so that discoveries resulted from an organized systematic process, as well as individual brilliance.  One needs both, of course, and yet anything worth knowing in science and technology needs to be written down and quantified, so that it can be further understood and improved. In thew words of William Thomson, Lord Kelvin, “When you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind.”  That’s the point: If you want anything to last, it has to expressed in a way that others can understand; and math is the most universal of languages. 

Shifting disciplines again, we can look to the Yale historian Edmund S. Morgan, who said that good scholar needed the “courage to simplify.”  Echoing Ockham, Morgan continued,“To simplify where you know little is easy.  To simplify where you know a great deal requires gifts of a different order: unusual penetration of mind and, above all, sheer nerve.”

We might begin a deeper analysis by making an Allan Rothian point: In politics, the fundamental unit is the vote.  Indeed, as we shall see, if the focus is kept on votes, then we need to question everything that is not voting.  By putting the focus on the fundamental unit in baseball, Beane & Co. changed the baseball; by putting the focus on the fundamental unit in politics, maybe  we can change politics.

And in politics, if one thinks about how campaigns judge themselves, one might think about the numbers being tossed around to show that the campaign is doing well—money raised, social media impressions delivered, contacts made, money raised, etc.  And, of course, polls, even if hey are only are, as is said, snapshot of the moment.  Of course, what really matters to a campaign, in the end, is votes.  That is, Did your candidate get more votes than the other candidate?  Did the campaign win?  

Let’s drill down, imagining it’s our campaign.  Let’s start with the realization that most voters, in both parties, have already resolved how they’re going to vote.  This is the base we hear so much about: those voters who vote in every election, or almost every election, always, or almost always, for the same party.  Probably 80 percent of voters count as base voters, roughly split between the two parties.  So if a base voter is one of “ours,” here’s our opportunity to insert some Moneyball-type thinking, by working to push the probability that he or she actually votes for our candidate to the highest possible level—ideally, of course, to the certainty of 100 percent.  Just as winning baseball teams strive to clear the bases by getting all their on-base players to home plate for the runs, so, too, we wish to “clear the bases” by getting all our voters to the voting booth (including, of course, mailboxes and any other place that receives ballots) for the election.  By this reckoning, politics, like baseball, is a numbers game: Get more runs to win; get more votes to win. 

So how, exactly, to get all our peeps to the polls?  This effort might entail any degree of of “white glove” treatment for the voter: nudges, reminders, peer pressure, made all the easier by big data and AI—imagine every vote-intending voter with a chatbot nearby.  Some might say that this is just good ol’ GOTV—Get Out The Vote—updated with new technology.  And in a way, that’s true: It is GOTV, made all the more granular with digital tools.  And for sure, there’s nothing devious or corrupt about it; our campaign is just applying the time-tested wisdom: Get there firstest with the mostest.  Yet of course, if this were easy, everyone would be doing it. 

Two Inefficiencies to be Ironed Out 

One inefficiency is on the vote side, the other on the elected official/party side.  As we think about this, we can keep this thought in the back of our minds: We can see that the costly superstructure of political campaigning is extrinsic.  What’s intrinsic, and needs to homed in on, is the actual process of getting voters to the voting booth—or, of course, these days, with absentee and mail-in voting, getting the voting booth to the voter.  Just as Roth wanted more runs, we want more votes.  Let’s keep it that simple.  Let’s talk about the voter first.

The Two-Screen Dilemma for Voters

Right now, the voter has to deal with two screens. 

The first screen is the one by which he or she sees the world of political news: typically either a TV screen or a computer screen, including the screen on a smart phone.  This is typically a news portal of some kind, be it cable news or a a news or commentary site.   This site, or sites, might be entertaining, but in general it is not politically interactive.  That  is, one can absorb information—video/audio, print—and yet one can’t typically take action on that site.  They are, in this sense, passively nontransactinal. 

The second screen is the one in which he or she takes action.  This would be more like a party website, or a donor platform such as Act Blue or WinRed. These are plenty interactive, but they aren’t, as a rule, entertaining,  That’s not their purpose: The purpose is to facilitate an action.  They are actively transactional.  Come here, give money, and perhaps do other things, such as apply for an absentee ballot.  And likely some day soon, vote.  

But here’s the question: Why are there two screens?   Everything in our world of consumer convenience says, unify, simplify.  That’s why we can put just about everything on our smart phone or tablet, from communication to premises security to tax payments.  The two-screen status quo flunks the Drucker Test: If we weren’t doing this now, would we start?  

So the voter is ill-served by this inefficiency.  But the politician, too, is ill-served.

The Opaque Screen for Politicians 

As a general rule, politicians like to talk.  But as a matter of campaign efficiency, their political advisers  should wish to talk first about their own election or re-election.  And here’s where we say where the media aren’t aligned.  After all, reporters and other kinds of journos want to talk about all the issues of the day, whatever they might be.  And so they will inevitably drag the pol into discussions of anything and everything.  The talented pol might well seek to steer the discussion to some stronger ground or safer harbor, but that’s not always possible.   It’s the rare pol who has the clout to demand that the topics raised on the segment or show stick to an approved menu.  

Moreover, to no small extent, cable news encourages a kind of circus-like atmosphere, since TV news is designed to entertain, as well as inform.  The pol might not be mind being an entertainer, but at some point, entertainment values get in the way of message-delivering.  There’s a reason why brands are careful about where their brand finds itself; pols, too, are a brand, and so they must be careful—even if they aren’t always so careful.  

Moreover, because the pol is talking on someone else’s platform, the audience is probably not legible to the pol.  Suppose a pol goes on Fox News, thereby speaking to a few million folks.  That’s all great, but the pol doesn’t know exactly who he (or she) is talking to.  All the pol knows is that he was on for three minutes.  Even Fox doesn’t know exactly who was watching, because TV ratings measure households and their general demographics, not the exact name and details.  Moreover, the pol-on-Fox will quickly realize that a repeat performance—and the repeat after that, and after that—most likely means that he is talking to pretty much the same people.  So not only does he not know the identities, but it’s likely that the unknowns are themselves, reruns.  So a rerun pol talking to a rerun audience.  That’s inefficient.  To put that another way, the signal-to-noise ratio isn’t so favorable.  This issue becomes all the more acute if the pol is focused on a critical election in his home state or district, as by definition, the vast majority of the audience will be outside of the needed jurisdiction.   (There’s a reason why the most common figures on cable news either have national ambitions or safe districts; if so, then they’re happy to play the cable game.) 

However, if the inefficiency of campaigns is a problem, the uncertainty of the voters is another problem.  And we’ll turn to that next.