Two Greek thinkers examine a secured firearm case, identification card and transfer ledger.

Universal Background Checks

Federal law requires a check on the buyer only when a licensed dealer sells the gun; nineteen states and the District of Columbia also require one for private sales. On 10 June 2025 Representatives Mike Thompson, a Democrat, and Brian Fitzpatrick, a Republican, reintroduced the bill. Most Republicans and gun rights groups oppose it, and it lies with the House Judiciary Committee.

22 percent of recent gun acquisitions in a national survey of owners involved no background check ^[4]
17,927 people were killed in firearm homicides in 2023 ^[1]
27,300 people died by firearm suicide in 2023, a record for the third year running ^[2]

The proposal

Almost every private gun sale goes through a licensed dealer who runs the check, with gifts within the family, loans at a range and hunting trips exempt.

The debate

Supporters say that felons and other prohibited buyers avoid the check by turning to private sellers, and that the form owners already fill in at a store costs little. Opponents answer that criminals get their guns illegally anyway, that the bill turns ordinary transfers between owners into crimes, and that California's law showed no effect. The House passed the bill on 11 March 2021, almost entirely along party lines, and the Senate never voted on it. The current version has not come to a vote.

“What the bill does do is close the private gun sale loophole, which has made it easy for felons and other prohibited purchasers to buy a gun online, at gun shows, or in person-to-person sales.”

argues for: Mike Thompson, Democrat, California House floor, 10 March 2021
“This law affects not only sales but any transfer of a weapon for any period of time. A couple of years ago, a 10-year study by Johns Hopkins and UC Davis concluded that California’s background check law had no effect on gun homicides or suicides.”

argues against: Tom McClintock, Republican, California House floor, 10 March 2021

The analysis

Universal Background Checks

90% of the weight is for, 10% against.

An argument's weight combines three questions: how big the effect is, how certain it is and how much the good at stake matters. The larger the slice, the more the argument weighs.

For

Against

Universal background checks are a good proposal on the current analysis: the costs are a dealer's fee, an afternoon of two people's time and a lightly narrowed liberty, while the benefit is measured in lives.

Very likely better

Very likely better: Under the assumptions used, the proposal brings more than it costs in 96 of 100 model runs.

Overall effect: large

How strongly the proposal works overall, in five steps from very small to very large, comparable across all debates. The ring is always full, even when the overall effect is small; this step says how large it is.

Thompson's side wins in the analysis too, though on thinner evidence than it claims. The result rests on an assumed small drop in homicide in the states without such a rule. That drop is set at a quarter of what studies of permit laws find, because the evidence for private-sale checks alone is limited. Against it stand many small items, from the dealer fee and the trip to the counter to the narrowed liberty of owners, together far lighter than the lives saved. McClintock made the opponents' strongest point: a careful study of California over a decade found no effect, and if private-sale checks change nothing, the measure is a pure cost.

In detail

All arguments

Arguments for3 arguments

People who are not shot

26points

There were 17,927 firearm homicides in 2023 and about 85,000 non-fatal firearm assaults. Roughly two thirds of both happen in the 31 states where a private sale needs no check at all.

Value 9.5 · Life and healthImpact 5.0Plausibility 5.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is people who are alive, and people who are not in a hospital with a gunshot wound, at the end of the period. It sits just below the top of the scale because it mixes deaths with a larger number of serious injuries, weighted towards the deaths, which is how this site prices a stream containing both. A tenth is added on the same anchor for the families of those killed and of those seriously injured, following the same rule used wherever this site counts violent death or serious injury. Nothing is added for fear, for the neighbourhoods where shootings concentrate, or for the medical spending avoided, each of which is real and each of which would enlarge this argument. Nothing is counted for the offenders. The value sits just below the maximum: the stream mixes deaths with a larger number of injuries, weighted towards the deaths.

American firearm homicides numbered 17,927 in 2023 and non-fatal firearm assaults about 85,000 ^[1]^[2]. Nineteen states and the District of Columbia already require checks on private sales, so the bill is new in 31 states holding roughly 65 percent of both figures: about 11,650 deaths and 55,000 injuries. The reduction used is four percent, in a range from one to twelve percent wherever the requirement bites at all; whether it bites at all is the separate question carried by the plausibility. That figure is set low deliberately: the research synthesis rates the evidence for private-sale checks as limited, and the larger effects in the literature come from permit-to-purchase laws, which add a licence and an in-person application that this bill does not ^[3]. Four percent gives 466 deaths and 2,210 serious injuries avoided a year. With a tenth added for the families of those killed and of those injured, that is a Normalised Impact of about 5; for comparison, the same score would stand for about 1.1 billion dollars a year of real money. The Impact is the largest here and it is a single assumed percentage away from being nothing at all.

▸ Show calculation ▾ Hide calculation

	Firearm homicides in the 31 states without a private-sale requirement [2]	17,927 nationally, about 65 percent of them	11,650 deaths a year
×	Reduction once private sales need a check Setting, range 1 to 12 percent wherever the requirement bites at all: a quarter of what permit-to-purchase studies report, because the synthesis rates the evidence for private-sale checks alone as limited [3]	4 %	466 deaths a year
×	Value of a life, with a tenth added for the families the value of a statistical life used across this site and the surcharge this site applies to violent death	1.4 million euro-equivalent each, plus 10 percent	718 million euro-equivalent a year
+	Serious injuries avoided, valued alongside them the value of a serious injury used across this site [1]	55,000 firearm assaults, 4 percent fewer, at 120,000 euro-equivalent each, plus a tenth for the families	292 million euro-equivalent a year
=	Deaths and serious injuries avoided, together		1,010 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	5.05

Score 5.05 Impact × 9.5 Value × 5.5 Plausibility ÷ 10 = 26 of 100

The counterfactual is the same states without the requirement, which is what the state-panel studies behind the research synthesis construct from the timing of state laws. Those designs are quasi-experimental — differences in differences across states and years — and the synthesis grades the resulting evidence for comprehensive checks on total homicide as moderate, its second-highest category ^[3]. The confounder that matters is that states legislating background checks also legislate everything else at the same time, from permit requirements to waiting periods to storage laws, so the effect attributed to checks may belong to a package; the strongest individual results in the field, from Connecticut's permit law and Missouri's repeal of one, are results about permits and not about checks alone. That is named and unresolved, and it is the reason the effect size used here is a quarter of what those studies report. Reverse causation is present in the raw correlation, since states with more violence legislate more, and the panel designs address it with state fixed effects. The Plausibility is a little above the middle: the direction is supported by a graded synthesis and the specific policy in this bill is the weakest-evidenced version of it.

Evidence: Converging studies · plausibility at most 6 Evidence of effect: Quasi-experimental · at most 8

Guns bought to die with

10points

Firearm suicides reached 27,300 in 2023, a record for the third year running and 58 percent of all gun deaths. A share of the guns used are bought shortly beforehand, and a check is a delay as well as a filter.

Value 9.5 · LifeImpact 2.7Plausibility 4

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is people alive at the end of the period who would otherwise have died by their own hand. It is placed just below the top of the scale, with a tenth added on the same anchor for the families, exactly as for the homicides above. It is a separate argument from those because the mechanism is different: a homicide is prevented by keeping a gun from someone prohibited, a suicide by putting a step and a delay between an intention and a method. Nothing is counted for attempts survived, which are numerous and would enlarge this. Nothing is counted for substitution to another method, which belongs in the quantity and is discussed there. The value sits just below the maximum, and this argument is separate from the homicides because a delay and a filter are not the same mechanism.

Firearm suicides reached 27,300 in 2023, 58 percent of all gun deaths, and about 65 percent of them occur in the 31 states this bill would newly reach — roughly 17,750 ^[2]. The reduction used is two percent, in a range from half a percent to six percent wherever the requirement bites at all, which is half the figure used for homicides. It is lower for a reason that runs against the argument: most people who die by firearm suicide use a gun that has been in the household for years, so a rule about acquisition reaches only the minority who buy shortly beforehand. That gives about 355 deaths a year, valued at the figure this site uses for a life with a tenth added for the families: a Normalised Impact of about 2.7 on the scale of this evaluation. Substitution is not subtracted separately, because the studies this rests on measure total suicide rather than firearm suicide alone, and a method that is substituted is already inside their estimate. The Impact is a bit over half the homicide argument, and it is smaller because most of the guns involved were bought long before the decision to use them.

▸ Show calculation ▾ Hide calculation

	Firearm suicides in the 31 states without a private-sale requirement [2]	27,300 nationally, about 65 percent of them	17,750 deaths a year
×	Reduction once private sales need a check Setting, range 0.5 to 6 percent wherever the requirement bites at all: half the homicide figure, because most firearm suicides use a gun that has been in the household for years [3]	2 %	355 deaths a year
×	Value of a life, with a tenth added for the families the value of a statistical life used across this site and the surcharge this site applies to violent death	1.4 million euro-equivalent each, plus 10 percent	547 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	2.74

Score 2.74 Impact × 9.5 Value × 4 Plausibility ÷ 10 = 10 of 100

The direction has support and the specific claim has a finding against it. The counterfactual is the same states without the requirement. Studies of comprehensive check regimes report reductions in firearm suicide, but the research synthesis rates that evidence as inconclusive and at least one careful state study — California's, examined over ten years after its comprehensive law — found no change in firearm suicide or homicide at all ^[3]^[6]. That is a counter-finding on the link that carries the quantity, not merely an absence of evidence, which is what places this argument where it sits rather than at the middle. The confounder is the same package problem as for homicides, and the direction check is the same: states with higher suicide rates legislate differently, and panel designs address it with fixed effects. Reverse causation is not otherwise plausible. The counter-finding is genuine, and the fact that the substitution mechanism is well documented in the suicide literature makes it easier to believe. The Plausibility is below the middle: the direction is supported in some states, a careful study of the largest state found nothing, and the mechanism reaches only the guns bought shortly before.

Evidence: Converging studies · plausibility at most 5 Evidence of effect: Quasi-experimental · at most 8 Placement: Load-bearing link disputed · plausibility 3.5–4

Open: Point-of-sale records and death certificates are both dated. Measuring what share of firearm suicides used a gun acquired within a year would settle how much of this mechanism can exist at all, and could move this argument in either direction.

Guns that cross state lines

2points

A state that requires checks on private sales still receives guns from the states next to it that do not. About three in ten guns recovered in crime in the strict states were first sold in another one.

Value 9.5 · Life and healthImpact 0.4Plausibility 5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is the same one as the first argument — people not killed — but in a different place: the nineteen states and the District that already require checks and still receive guns from the states that do not. It carries the same value class and the same family surcharge. It is separate rather than folded in because the first argument counts only the states where the rule is new, and this is the effect the rule has where it already exists. Nothing is counted for the traffickers. Nothing is counted for the strict states' enforcement costs, which fall. The value sits just below the maximum for the same reason as the first argument, and this stands apart because it lands in the states the first argument leaves out.

About 6,275 firearm homicides a year occur in the nineteen states and the District that already require checks on private sales. Federal tracing data show that roughly three in ten guns recovered in crime in such states were first sold in another state, and the source states are overwhelmingly the ones with the loosest transfer rules. A national requirement closes the supply at its origin rather than at the border. The reduction used is three percent of the out-of-state share, in a range from one to ten percent wherever a national requirement changes the flow at all — about 56 deaths a year, a Normalised Impact of 0.43 with the family surcharge. The figure is small because trafficking survives a paper requirement: a straw purchaser who passes a check buys the gun legally and hands it on, which no check catches. The Impact is a tenth of the first argument, which is the honest size of an effect that works on the supply of trafficked guns rather than on the buyer in front of the counter.

▸ Show calculation ▾ Hide calculation

	Firearm homicides in the states that already require checks [2]	17,927 nationally, about 35 percent of them	6,275 deaths a year
×	Share involving a gun first sold in another state federal tracing data for states with strict transfer rules [5]	30 %	1,883 deaths a year
×	Reduction once the source states also require checks Setting, range 1 to 10 percent wherever a national requirement changes the flow at all: straw purchasing defeats a check entirely for anyone using an accomplice with a clean record	3 %	56.5 deaths a year
×	Value of a life, with a tenth added for the families the value of a statistical life used across this site and the surcharge this site applies to violent death	1.4 million euro-equivalent each, plus 10 percent	87 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	0.43

Score 0.43 Impact × 9.5 Value × 5 Plausibility ÷ 10 = 2 of 100

The counterfactual is the strict states as they are, receiving guns from loose ones. The chain is named and each link is documented separately: tracing data establish the interstate flow, and the concentration of source states with weak transfer laws is a repeated finding in the tracing literature. What is not measured is the last link — how much of that flow a national requirement would stop — because no national requirement has existed to measure. The confounder that matters is straw purchasing, which defeats the mechanism entirely for anyone willing to use an accomplice with a clean record; it is named and unresolved and is the reason the effect is set at three percent rather than at the whole out-of-state share. Reverse causation does not arise. The Plausibility is at the middle: the flow is measured, the share of it a check requirement would stop is assumed, and straw purchasing limits it by construction.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6

Arguments against4 arguments

Handing a gun to a friend

1.8points

The bill turns an unrecorded private transfer into a federal offence unless it goes through a dealer or fits an exemption. Forty million gun owners in the affected states would be living under a rule about something they have always done informally.

Value 9 · Rights and libertyImpact 0.4Plausibility 5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is a liberty narrowed: the freedom to hand a rifle to a neighbour, sell a shotgun to somebody at a club, or pass a pistol to a nephew without involving a licensed intermediary. This site places that at the constitutional core, because what is restricted is the exercise of a right the Supreme Court has held to be individual, and the restriction is a criminal one. What is priced is the restriction itself, spread evenly across the owners it applies to, and not the disapproval of people who dislike the law. The prosecutions that follow are few and are inside the same figure rather than a separate one. Nothing is counted for the exemptions in the bill, which cover gifts to close family, loans at a range and temporary hunting transfers, and which are why the figure per owner is small. The value is at the constitutional core because what is narrowed is the exercise of an individual right under criminal penalty, and the amount per owner is small because the ordinary cases are exempt.

About 40 million adults own a firearm in the 31 states this bill would newly reach. The restriction on them is real and it is not large, because the transfers people actually make most often — to a spouse, a parent, a child, a friend at a range, a hunting companion for the day — are exempt on the face of the bill. What is left is the sale or lasting loan to someone outside that circle, which most owners do rarely or never. The restriction is priced as if it cost each owner about two dollars a year, in a range from fifty cents to nine dollars, which is the band this site uses for a restriction that applies broadly and bites lightly; the same band prices the loss of anonymity in the platform evaluations elsewhere on this site. That gives a Normalised Impact of 0.4 on the scale of this evaluation. Prosecutions are inside this figure: they run in the low hundreds a year in states with such laws, and a case is worth far more than that to the person facing it, but there are too few of them to move a figure built on 40 million people. The Impact is a twelfth of the lives on the other side, and it is built on a per-person figure rather than on any measured harm.

▸ Show calculation ▾ Hide calculation

	Firearm owners in the states the bill newly reaches [4]		40 million adults
×	Value of the restriction per owner per year Setting, range 0.5 to 8 euro-equivalent (about 0.55 to 9 dollars): the band this site uses for a restriction that applies broadly and bites lightly, since transfers within the family, at a range and for hunting are exempt	2 euro-equivalent (about 2.20 dollars)	80 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	0.4

Score 0.4 Impact × 9 Value × 5 Plausibility ÷ 10 = 1.8 of 100

The restriction is definitional — the statute says what may not be done — but what it is worth to the people it applies to is a valuation and not a finding. The counterfactual is the same owners under present law in those states. The chain is short and complete: a transfer that is lawful becomes an offence, and the people who would have made it lose the option. The confounder is the exemptions: how broadly a court reads the family, range and hunting carve-outs decides whether this costs an owner almost nothing or a good deal more, and it is unresolved. Reverse causation does not arise. The per-owner figure is the site's own convention for a broad, light restriction rather than anything measured about firearm owners, and if the restriction is felt more sharply than that convention assumes, this argument is several times larger. The Plausibility is at the middle: the restriction is certain and its value to those it binds is a convention.

Evidence: Plausibility · plausibility at most 5 Evidence of effect: Definitional

Paying for the paperwork

1.5points

Every private sale would have to go through a licensed dealer, who charges a fee for the service — about 33 dollars on every transfer — and the federal system has to run the additional checks. The time it takes the buyer and the seller is counted separately.

Value 5 · Enforcement costImpact 0.4Plausibility 7

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is money consumed in complying with the rule: the dealer's transfer fee and the federal system's cost of running the additional checks. It is priced at the middle of the scale, like other money spent on running a rule. Both halves are real costs rather than money changing hands: the fee buys the staff minutes and the record-keeping that a transfer takes at the counter, which is why it is not set against the dealer's receipt, and the federal cost buys examiners and systems. The buyer's and the seller's own time is a different good — hours of a person's life given up because a rule requires it — and this site prices it in a different class, so it stands as its own argument below. Nothing is counted for the transfers that do not happen because of the friction, which would be a further loss and is not estimated. The value is the middle of the scale, the level this site uses for the cost of running a rule.

Private firearm transfers in the United States run somewhere near four million a year, of which about 2.6 million occur in the states this bill would newly reach ^[4]. Each would cost a dealer's transfer fee, which runs from about 25 to 50 dollars; about 33 dollars a transfer is used, or 87 million dollars a year. On top of that the federal system has to process the additional checks: at about 3.30 dollars a check, the running cost of the system spread over its volume, 2.6 million more checks cost about 9 million dollars a year ^[7]. Together that is about 96 million dollars a year at the ordinary weight for money. The fee would be larger in rural counties where dealers are few, which is the substance of the objection made by the bill's opponents and is inside the range rather than outside it. The Impact is a twelfth of the lives saved on the other side, which is the ordinary proportion when a rule is cheap per transaction and the harm it addresses is fatal.

▸ Show calculation ▾ Hide calculation

	Private firearm transfers in the states the bill newly reaches [4]	about four million a year nationally, 65 percent of them	2.6 million transfers a year
×	Dealer transfer fee Setting, range 22 to 45 euro-equivalent a transfer: a dealer fee of 25 to 50 dollars, converted at 0.90 euro to the dollar (range 0.85 to 0.95)	30 euro-equivalent (about 33 dollars)	78 million euro-equivalent a year
+	Plus the federal cost of running the additional checks the running cost of the federal system spread over its volume [7]	2.6 million checks at about 3 euro-equivalent each (about 3.30 dollars)	86 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	0.43

Score 0.43 Impact × 5 Value × 7 Plausibility ÷ 10 = 1.5 of 100

The costs here are prices rather than predictions. The counterfactual is the same transfers taking place privately at no cost. Dealer transfer fees are published and competitive, the federal system publishes its volumes, and the number of private transfers is estimated from owner surveys rather than counted, because private sales are by definition unrecorded — that survey estimate is the softer half of the calculation and it is wide. The chain has no behavioural link that carries the quantity: if the rule is followed, the fee is paid and the check is run. The confounder is compliance in the other direction, which would lower this cost and lower the benefit at the same time, so it does not move the balance. Reverse causation does not arise. The Plausibility is high: the fee is a published price and only the number of private transfers is estimated.

Evidence: Precedent · plausibility at most 8 Evidence of effect: Definitional

A trip to the dealer

0.88points

A private sale that once took place at a kitchen table becomes a drive to a licensed dealer and a wait while the check runs — for both parties, on every transfer. That is time out of people's lives, and there are 2.6 million transfers a year.

Value 9 · LifetimeImpact 0.1Plausibility 7

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is hours of people's lives given up because a rule requires it: the drive to a dealer, the wait at the counter, the drive back, for the seller and the buyer alike. This site treats time somebody has to spend as the same good as the years of life it counts elsewhere, derived from the same anchor, and it places it in the class it uses for life and health — a step below the top, because an afternoon is recoverable in a way a life is not. It is priced at the full value of a waking hour rather than the lower rate for freely chosen travel, because nothing else can be done with the time and nothing is gained from it. It stands apart from the fee, which is money and sits in the class for the cost of running a rule. The value is just below the top of the scale because the stream is lifetime taken under compulsion, priced at the full rate for an hour with nothing else in it.

About 2.6 million private transfers a year would have to go through a dealer in the states this bill newly reaches ^[4]. Each takes about an hour and a half of the two parties' time between them, in a range from one to three hours — the drive, the wait while the check runs, the paperwork — and the upper end is the rural county where the nearest dealer is an hour away. That is about 3.9 million hours a year, valued at the full rate this site uses for a waking hour a rule takes without giving anything back, close to an hour's pay at the federal minimum wage: a Normalised Impact of 0.14 on the scale of this evaluation. Nothing is counted for transfers that are delayed by a check that does not clear at once, which adds a second trip for a minority. The Impact is a third of the fee it accompanies, and it is priced as time taken from people rather than as money spent.

▸ Show calculation ▾ Hide calculation

	Private firearm transfers in the states the bill newly reaches [4]	about four million a year nationally, 65 percent of them	2.6 million transfers a year
×	Hours of the two parties' time per transfer Setting, range 1 to 3 hours: the drive, the wait while the check runs and the paperwork; the upper end is the rural county where the nearest dealer is an hour away	1.5 hours	3.9 million hours a year
×	Value of a waking hour taken under compulsion the full value of a waking hour this site uses for time a rule takes without giving anything back	6.85 euro-equivalent (about 7.60 dollars)	27 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	0.14

Score 0.14 Impact × 9 Value × 7 Plausibility ÷ 10 = 0.9 of 100

The time is a consequence of the rule rather than a prediction about behaviour. The counterfactual is the same transfers taking place privately, with no journey and no wait. What is estimated is the number of private transfers, which comes from owner surveys because private sales leave no record, and the hours per transfer, which are a setting within a range. The chain has no behavioural link carrying the quantity: if the rule is followed, the trip is made. The confounder is non-compliance, which would lower this cost and the benefit together and does not move the balance. Reverse causation does not arise. The Plausibility is high: the trip is what the rule requires, and only the number of transfers and the hours each takes are estimated.

Evidence: Precedent · plausibility at most 8 Evidence of effect: Definitional

Checks that stop the wrong people

0.22points

The federal system returns a wrong denial in a small share of cases, and the appeal takes months. About 2.6 million additional checks a year means some fifteen hundred more lawful buyers turned away.

Value 9 · Rights and libertyImpact 0.04Plausibility 6

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is the wrongly refused: people entitled to buy who are matched to somebody else's record, or to a disposition that was never updated, and who wait months for it to be corrected. This site places that at the constitutional core, because what is denied is the exercise of a right on the basis of an error by the state. Nothing is counted for the delay experienced by everyone else, which is inside the time argument. Nothing is counted for the correctly denied, who are the point of the measure. The system's running costs are money and sit with the fee, in the class this site uses for the cost of a rule rather than here. The value is at the constitutional core because what is at stake is a right denied by mistake, and the number of people it happens to is what keeps the argument small.

About 2.6 million additional checks a year would run under this bill in the states it newly reaches — the same volume that carries the fee on the previous argument. In 2023 the federal section processed 10.2 million checks and denied 116,587 of them, just over one percent; it received 22,615 challenges to denials and overturned 6,263, which is 5.4 percent of all denials, most often because a court disposition had never been recorded ^[7]. Applied to the additional volume, that is about 1,500 wrongly refused buyers a year. What a wrong refusal costs is priced at the band this site uses for an unfounded suspicion recorded against a named person, which is what a denial is: as if each cost the person about 5,600 dollars, in a range from 1,100 to 22,000. That gives a Normalised Impact of 0.04 on the scale of this evaluation. Nothing is added for the days between a denial and its correction, which are inside the same band. The Impact is the smallest here and it is the argument most likely to be understated, because the error rate is measured only at the point of appeal and most people do not appeal.

▸ Show calculation ▾ Hide calculation

	Additional checks a year in the states the bill newly reaches [4]	the same 2.6 million transfers that carry the fee	2.6 million checks
×	Denials that are later overturned federal counts for 2023: 116,587 denials among 10.2 million checks, 6,263 overturned on challenge [7]	1.1 percent denied, 5.4 percent of those overturned	1,500 people a year
×	Cost of a wrong refusal Setting, range 1,000 to 20,000 euro-equivalent (about 1,100 to 22,000 dollars): this site's band for an unfounded suspicion recorded against a named person	5,000 euro-equivalent (about 5,600 dollars)	7.5 million euro-equivalent a year
÷	Normalised Impact scale of this evaluation	200 million euro-equivalent a point	0.04

Score 0.04 Impact × 9 Value × 6 Plausibility ÷ 10 = 0.2 of 100

The counterfactual is the same buyers under the present system, in which private sales are not checked at all. Both inputs are administrative counts rather than estimates: the federal system publishes denial volumes and the outcome of every challenge ^[7]. The chain is short — more checks, proportionally more denials, proportionally more errors — and it has no behavioural link carrying the quantity. The confounder is that the overturn count is not the error count: 6,263 overturns among 116,587 denials is the share corrected, and somebody wrongly denied who does not challenge never appears in it, which means the true number is higher than the one used and this argument is understated rather than overstated. That is named and points against the direction of the setting. Reverse causation does not arise. The Plausibility is a little above the middle: the counts are administrative and the error rate is measured only among those who appeal.

Evidence: Precedent · plausibility at most 7 Evidence of effect: Definitional

How robust the result is

Under the assumptions used, the proposal brings more than it costs in 96 of 100 model runs.

gains more than it costscosts more than it gains

Each dot stands for 200 of the 20,000 model runs.

Everything on the positive side rests on one assumed number: that closing the private-sale gap lowers firearm homicide by four percent and firearm suicide by two percent in the states it newly reaches. Both figures are set low on purpose, at a quarter of what permit-to-purchase studies report, because the research synthesis rates the evidence for private-sale checks alone as limited ^[3]. With them the proposal comes out ahead in 96 of 100 runs. At zero effect the measure is a pure cost and the balance turns negative. At the effect sizes found for permit laws, three to five times these, it would be the strongest measure on this site. The lightly narrowed liberty of forty million owners is this site's convention for a broad restriction that bites lightly, not anything measured about gun owners; felt more sharply, that argument would be several times larger ^[4]^[5].

Show the distribution of the model runs

20,000 simulated futures

Very likely better · 96 %

In 96 of 100 simulated futures the proposal brings more than it costs.

No effect is known exactly; every argument has a range. The calculation draws a possible size for each argument from its range 20,000 times and counts how often the benefits exceed the costs.

Net effect in points per future. Left of zero the drawbacks outweigh the benefits, right of zero the benefits win. Eight in ten futures fall between +2.8 and +79 points.

Scale and assessment

Scale of this evaluation: Normalised Impact, unitless and calibrated to this topic. For comparison, one point here corresponds to about 200 million euro-equivalent a year.

Size class:medium

Effect large (+6.9) · 90 % of the effect is positive

Areas show individual argument weights. The positive share of the effect nets out money transfers between the two sides; the probability of a positive net effect is determined separately.

How we score →

Review and sources

AI evaluation · not yet reviewed by a human

This evaluation was produced and sourced by an AI model; a human review is still pending. Figures and conclusions may still change. The review log is at the foot of the page.How review works →

Last reviewed by Claude Opus 5.5 · October 29, 2026 · 7× AI, not yet reviewed by a human

October 29, 2026AI reviewClaude Opus 5.5record updated
Page texts for the new page layout (draft G) with quotes from primary sources; numbers unchanged.
October 21, 2026AI reviewClaude Fable 5.1record updated
Money amounts in argument texts stated in dollars per style-en.md; scores, chains and balance unchanged
October 20, 2026AI reviewClaude Fable 5.1record updated
Summary and scale note rewritten to the style of 20 Oct 2026; numbers unchanged
October 14, 2026AI reviewClaude Opus 5.5record updated
Prüfbericht 11.10.: Unterüberschrift „almost every“; pro-1 Wirkung als Todesfälle, Verletzte und Normalised Impact statt Eurobetrag (G5), Zettel mit eigener Summenzeile (G2).
September 14, 2026AI reviewClaude Opus 5re-scored
Gesamtprüfung 08.09.: Castillo-Carniglia-Quelle korrigiert, Kostenmenge einheitlich 2,6 Mio (con-3 I 0,022→0,0075), Zeit als eigenes con-4 (V 9, I 0,027), Gebühr+Bundeskosten V 5 (I 0,086), Angehörigenzuschlag für Verletzte (pro-1 I 0,983→1,01); r 0,90 → 0,90.
September 6, 2026AI reviewClaude Opus 5record updated
i_spanne an allen 6, normalisierung an 5 von 6 (leben, verletzung-schwer, personenstunde, eur-nennwert; zwei Grundrechts-Stroeme ohne Anker), Ob-Gruppe checks-change-who-gets-a-gun ueber die drei Pro-Argumente, kein Transfer im Record. Zwei gesetzte Konventionen: Dollar-Umrechnung 0,90 EUR/$ (Spanne 0,85-0,95, die Kette rechnete mit 1,20) und Haendlergebuehr als realer Ressourcenaufwand. Kategorie kippt von Deutlich besser (r 0,90) auf Ausgeglichen (P(D>0) 0,56).
September 6, 2026AI reviewClaude Opus 5First evaluation
First evaluation: effect sizes set at a quarter of what permit-law studies report, because the synthesis rates private-sale checks alone as limited.

Evaluations are produced with AI support and reviewed on a schedule for new developments; human passes are marked separately.How we review →

Show all 7 sourcesHide sources

Centers for Disease Control and Prevention: Firearm injury and death: fast facts. cdc.gov
Johns Hopkins Center for Gun Violence Solutions: Gun violence in the United States 2023. publichealth.jhu.edu
RAND Corporation: The effects of background checks, in The Science of Gun Policy. rand.org
Miller, Hepburn and Azrael, Annals of Internal Medicine: Firearm acquisition without background checks: results of a national survey. acpjournals.org
Bureau of Alcohol, Tobacco, Firearms and Explosives, and the Federal Bureau of Investigation: Firearms trace data and National Instant Criminal Background Check System operations reports. atf.gov
Castillo-Carniglia and colleagues, Annals of Epidemiology: California's comprehensive background check and misdemeanour violence prohibition policies and firearm mortality (Annals of Epidemiology 30, 2019, 50–56). pubmed.ncbi.nlm.nih.gov
Federal Bureau of Investigation, Criminal Justice Information Services Division: National Instant Criminal Background Check System (NICS) 2023 Operations Report. fbi.gov

More debates in this area

Very likely better · 97 %

Size: small

Federal Red Flag Orders

Create a federal court order that temporarily removes firearms from a person shown to be a danger to themselves or others, available in every state rather than only the twenty-two that have their own.

All debates in Guns and Public Safety →