There’s a moment in Sebastian Junger’s excellent book War (2010) that has stuck with me since I read it. It’s when Junger writes about watching American soldiers in Afghanistan’s Korengal Valley firing weapons that cost tends of thousands of dollars from helicopters and planes that cost millions of dollars to kill a Taliban sniper who was earning between five and ten dollars a day. Part of the disparity is systemic: the weapons the Americans had to deploy were not designed or built for fighting insurgents, but for destroying Soviet tanks in Germany’s Fulda Gap. “You go to war with the army you have,” as Donald Rumsfeld observed - which is fine when the war it is a war of necessity (Poland in 1939 sending cavalry against the mechanized Wehrmacht(, not when it is a war of choice.
But there’s a deeper phenomenon at play here: overkill, the expenditure of disproportionate resources to achieve a goal. Consider the following analogous situations:
Hillary Clinton’s 2020 election campaign chose, at the behest of supposedly the most advanced (certainly the most expensive) data analytics to buy costly television advertising exclusively during programs which are overwhelmingly watched by progressives highly likely to turn out and vote for whoever the Democratic candidate that year was.
In the mid-2000s Ed Catmull, the head of Pixar, noticed with alarm that the company’s films, while still critically acclaimed and commercially successful, were taking longer and longer to produce, and consequently costing more and more money. He didn’t have to investigate long to discover that Pixar’s animators were massively overproducing content: in one notable instance, from Monsters Inc. (2001), a stack of CDs falls over. The sequence is onscreen for less than a dozen frames, but the animators created custom CD graphics, cover art, and liner notes for every single CD, all of which was invisible to audiences.
A recurrent issue in software development and engineering happens when teams develop from scratch new tools and techniques, even to the point of inventing entirely new programming languages, all for the purposes of delivering business accounting software indistinguishable from the competition.
The United States’ early SIOPs (Single Integrated Operational Plan - the nation’s Top Secret plans for waging nuclear wars against the Communist bloc) called for dropping hundreds of nuclear weapons on Moscow. Studying the plans in the 1960s, Secretary of Defense Robert McNamara realized that the airspace over Moscow would be so crowded that B-52 bombers would be destroying each other with the electromagnetic pulses, shockwaves, and heat flashes from their respective bombs, all in the quest to make sure the Kremlin, the Lubyanka, and the Bolshoi were all reduced to finer and and more radioactive rubble.
University campuses blanket themselves with posters, informational sessions, Codes of Conduct signings, and other material all in the aim to prevent sexual assaults. This material overwhelmingly does not reach or sway the people conducting the sexual assaults - it changes no minds, and prevents few sexual assaults, as it is consumed primarily by those least likely to commit the crime.
Fashionable causes like climate policy advocacy, environmental and historic landmark preservation, Third World poverty relief, animal rights, and supplying social services to the destitute, homeless, and mentally ill attract a plethora of overlapping, redundant nonprofits that compete with each other for talent, for volunteers, and for funding.
An example from Robin Hanson: huge amounts of healthcare spending in the developed world is wasted - it does not make people healthier, and frequently causes iatrogenic harm (harmed caused by healing. The problem is well known, but persists.
Human society’s are highly prone to overkill. It happens from national governments down to families. Given the staggering waste involved, we need to ask: why does it happen?
Why Overkill?
Overkill happens for a number of overlapping reasons:
Shooting in the Dark: It’s an oft-cited statistic, used to justify the sky high price of novel pharmaceuticals, that bringing a drug to market costs on average >$1 billion, from initial research through to final regulatory approval, manufacturing, and distribution. Any examination of the accounting of a pharmaceutical company will show that this is not strictly true: considering the molecule/biologic under study by itself, the cost is closer to $50-100 million. However, the ratio of novel molecules investigated to drugs brought to market is something like 10 or 20 to 1. The new drug has to pay for all the dead ends the drug company pursued.
Here is probably where overkill is most justifiable: if you could go back to the 1960s and bring the molecular structure and manufacturing steps to make herceptin, ledipasvir, metformin, and rosuvastatin, it would not cost (inflation adjusted) $1 billion to bring those drugs to market.
Overkill can happen here and in similar domains, perhaps unavoidably, because the search space is very large and the starting point is uncertain. You may know the target - in pharma terms, the ‘lock’ you want to fit the molecular ‘key’ to - but the path from where you are to there is shrouded in darkness.Unknown Culminating Point of Victory: In military science and strategy, there’s a point in any offensive known as the “culminating point of victory,” the point where the maximum defensible advance has been made. Smart commanders stop the advance of their forces at this point, allowing a period for rest and replenishment. Passing it can lead to overextension, exhaustion of logistics, and vulnerability to enemy counterattacks. Knowing exactly where the culminating point is requires experience, skill, and self-control.
The world is complex,1 and many of the most intractable problems are opaque to inquiry: the information can be difficult to collect, the true state of the situation may only be known through proxy measures, or and it can be difficult to even figure out what measures you need to use. As a result, its difficult to know when the maximum good of the subsidy, tariff, or public health campaign has been reached.Lagging Indicators: A sub problem of an Unknown Culminating Point of Victory is a lagging indicator. You may have a good indicator for your progress, one that accurately reflects the state of the world you care about influencing, but it may take a long time for that indicator to become evident enough to be reported upon. I’ve heard it remarked that primary and secondary education is an intractable world to management by measurement because the product takes eighteen years to produce: is the boy a good citizen, law-abiding, gainfully employable? Tests can tell you whether he is learning how to take school and state-set tests successfully, not whether his knowledge of “fractions and gym” equips him to deal with real world situations.
Inverse Drucker: Peter Drucker, the mid-20th century management thinker, is credited with the saying “what gets measured gets managed,” more informally “what gets measured gets done.” When dealing with complicated situations, people and the organizations they compose have a tendency to invert this into “what is easy to do gets measured,” i.e. that measure which we can most visibly affect, we will choose as our measure of success. Horrifically, this was one of the metrics the United States used to measure its “progress” during the Vietnam War - in a war without front lines, no possible invasion of North Vietnam, and with severe restrictions on expansion then the more dead “Viet Cong” the ARVN and US military could produce, the closer victory must be.
“We Must (Be Seen to Do) Something!”: There’s a quote I like a lot, often attributed to President’s advising their staff not to engage in effort for the sake of effort: “don’t just do something, stand there!” I first heard it attributed to Herbert Hoover, advising his Secretary of Commerce to not engage in major interventions in the market to counter the 1929 Stock Market Crash and the ensuing Great Depression. Such, whether for good or ill, is rare sentiment from governments and the politicians who lead them.
The need to do something, to engage in action no matter how ill-informed, ill thought out, or futile, comes from both within and without. Everyone in a stressful situation who has soothed themselves by some activity, whether it is cleaning the bathroom or baking a cake, knows the internal need.2 Studies of post traumatic stress disorder in soldiers on active duty have noted that it far more commonly affects behind the lines logistics troops than front line troops: if you can shoot back at the enemy, you are less frightened and traumatized than you are if your convoy is suddenly turned into flames and shrapnel by a roadside bomb.
The external need is the belief, rightly or wrongly, that other people expect you to contribute to solving the situation. For difficult, intractable issues like drug abuse, governments are held responsible by their populations, and expected to do something about it. Of course, public attention is on what they can see, not on what they cannot see (indeed, as Bastiat remarked in one of the greatest philosophical essays ever,3 that which is seen prevails in men’s minds so much over what is unseen that they hardly ever imagine there is more to know). Therefore, it is in the interest of government to take visible actions against perceived problems, even when less visible actions may be more successful.4X Is Not About Y - Growing out of We Must (Be Seen To) Do Something, is when the effort expended is not about the stated goal at all. The work may still continue, but the expenditure of time, effort, and money is about something else. It is Robin Hanson’s claim that the modern medical system, outside of vaccinations, antibiotics, and emergency medical care, is really about showing care. Bryan Caplan makes a similar claim about education, that the skills taught, beyond literacy and numeracy, are beside the point; the real point of education is for students to demonstrate self-discipline, intelligence, and submissiveness. If true, these are regrettable but understandable: they are not consciously done.
Less forgivable is when “X is Not About Y” is consciously understood by its practitioners to be the case: the purpose is not the stated goal, but the advancement of the reputations/careers/social standing of those involved. In Emily St. John Mandel’s Sea of Tranquility, the protagonists learn that the bureacuracy they serve primarily exists to perpetuate the bureaucracy. Philanthropy, except when undertaken in Christ’s spirit5, is primarily about the ego and social standing of the giver: observe that it is easier to attract donors to putting their names on static buildings than to funding operations, maintenance, and training. Any responsible fundraiser knows this, and knows that flattering the donors is an important part of their job: the more visible the perk, the more you can reasonably ask for it.
I have no positive message to end this essay with: overkill seems to be endemic and irremovable from human affairs. Readers are welcome to post their own examples of overkill from domains they are familiar with in the comments.
Citation not needed.
Incidentally, marital advice: if you are arguing with your spouse and things are getting heated, take a break and go and clean something. You’ll both feel better for having a cleaner place to live, and the effort expended shows your your spouse your commitment to them through your care of your mutual living space.
It’s an essay in political economy, but its point is much deeper than just observations on the effects of trade and tariffs.
Racism, systemic and otherwise, may have played a role, but I believe the need to take visible action explains the overwhelmingly supply-side focus of the so-called War on Drugs. Mass arrests and seizures of contraband are visible markers of ‘doing something’ in a way that safe injection sites, poverty relief, and offering narcotics counselling simply are not.
“Take heed that ye do not your alms before men, to be seen of them: otherwise ye have no reward of your Father which is in heaven. Therefore when thou doest thine alms, do not sound a trumpet before thee, as the hypocrites do in the synagogues and in the streets, that they may have glory of men. Verily I say unto you, They have their reward.
But when thou doest alms, let not thy left hand know what thy right hand doeth: That thine alms may be in secret: and thy Father which seeth in secret himself shall reward thee openly.” (Matthew 6: 1-4, KJV translation)