Today, the U.S. Consumer Product Safety Commission voted to issue a Notice of Proposed Rulemaking (NPR) to address blade-contact injuries among users of table saws. Unfortunately, I was unable to join in this decision.
The Risk and the Mitigation
As discussed in detail in the excellent briefing package assembled by our talented and dedicated staff, there is certainly a risk in using table saws. Tens of thousands of consumers a year[1] suffer injuries – ranging from relatively minor lacerations to finger amputations – from contact with the blades of table saws. The saws have blade guards that are intended to reduce the odds of such contact, but those guards are not and likely cannot be 100% effective. First, the guards must be removed to perform some advanced cuts. Second, a scenario called “kickback” (described in more detail below) can cause sudden blade contact for even a conscientious user. Third, some consumers simply remove the guard because they find it inconvenient or resent it being forced on them.
In the early 2000s, Dr. Steve Gass – an inventor and attorney with a doctorate in physics – developed a system he called SawStop, which used sensors to detect flesh in the blade path and quickly stop the blade.[2] The device’s operation may destroy the blade, but it saves the finger. SawStop and any similar devices to detect potential injury and stop or remove the blade are collectively known as Active Injury Mitigation technology, or AIM.[3]
Over a period of years, Dr. Gass negotiated with a variety of saw manufacturers regarding licensing his technology for their saws, but no agreements came from these negotiations. In 2003, Dr. Gass and his business partners petitioned CPSC to require AIM technology on table saws through a mandatory product safety standard under the Consumer Product Safety Act (CPSA). In 2004, SawStop removed the middleman and released its own saw.[4] Since then, another manufacturer, Bosch, has brought to market a saw with similar technology,[5] though SawStop has pursued patent enforcement against Bosch.[6] Each of these AIM-equipped saws addresses the hazard of blade contact by “remov[ing] the spinning blade from the point of contact quickly enough, within milliseconds, to reduce significantly the severity of injury.”[7]
There is no question, then, that a substantial number of consumers suffer horrific injuries using table saws. There is also no question that technology is available to consumers to dramatically reduce (though not eliminate) the risks of such injuries. So why did I vote against the rule? Precisely because technology is available to consumers that can dramatically reduce the risks of such injuries.
That is not to say that I would never vote to propose or issue a rule where existing technology can protect the consumer. Indeed, only very rarely would it be appropriate for CPSC to demand a level of performance we did not already know to be achievable. But, where a specific series of factors is present, I believe that the Commission should, as a matter of responsible government, that respects consumer choice and individual liberty, allow consumers to decide for themselves what risks they will tolerate and what protection against those risks they will demand.
What is an Unreasonable Risk?
The core of the CPSC’s mission is “to protect the public against unreasonable risks of injury associated with consumer products.”[8] The Consumer Product Safety Act (CPSA), which both created the agency and gave us that solemn duty, does not define an “unreasonable risk.” It impliedly leaves that task to each commissioner, to the Commission as a whole, and to the body politic – the president, Congress, the courts, and of course the American people.
The legislative history to the CPSA does provide some insight. The original Senate version of the bill that became the CPSA defined an unreasonable risk as one that the agency felt was “unwarranted because the degree of anticipated injury or the frequency of such injury can be reduced without affecting the performance or viability of the consumer product or . . . the effect on such performance is justified when measured against the degree of anticipated injury or the frequency of such injury.”[9]
When the bill went to conference, the Senate abandoned its definition and adopted the House’s choice to leave the term undefined.[10] The House’s thinking, however, mirrored the Senate’s closely.
"An unreasonable hazard is clearly one which can be prevented or reduced without affecting the product’s utility, cost, or availability; or one [for] which the effect on the product’s utility, cost, or availability is outweighed by the need to protect the public from the hazard associated with the product.[11]"
Courts did not take long to pick up this concept as they tried to define the boundaries of what the agency could do,[12] and it has become ingrained in the agency’s ethos. In Aqua Slide, for example, the Fifth Circuit held that the risk of “quadriplegia and paraplegia resulting from users . . . sliding down the slide in a head first position and striking the bottom of the pool,” while “extremely remote,” was nonetheless so severe that CPSC could regulate it as unreasonable if it came up with “a standard which actually promised to reduce the risk without unduly hampering the availability of the slides or decreasing their utility.”[13]
Beyond this balancing test, however, courts have recognized that consumers have their own role in determining what risks are unreasonable, and that the agency should only step in where something prevents them from playing that role.
"[A]n important predicate to Commission action is that consumers be unaware of either the severity, frequency, or ways of avoiding the risk. If consumers have accurate information, and still choose to incur the risk, then their judgment may well be reasonable.[14]"
One of CPSC’s original commissioners expressed a similar notion of consumers as their own protectors, though he added a few more qualifications.
"In a speech, Commissioner David Pittle stated his definition of a ‘reasonable risk’ as ‘one where a consumer (A) understands by way of adequate warning or by way of public knowledge that a risk is associated with the product; (B) understands the probability of occurrence of an injury; (C) understands the potential severity of such an injury; (D) has been told how to cope with the risk; (E) cannot obtain the same benefits in less risky ways at the same or less cost; (F) would not, if given a choice, pay additional cost to eliminate or reduce the danger; and (G) voluntarily accepts the risk to get the benefits of the product.[15]"
Under either of these frameworks, I believe the risk of injury from table saws that our NPR seeks to address is a reasonable one and not an appropriate subject for Commission action. I believe consumers have made a choice and, even if we would choose differently, we should respect that choice. Moreover, the market is testing whether or not consumers will choose a safer, albeit more expensive, product, just as Commissioner Pittle suggested they might, and I do not believe it is within our authority to make that choice for them.
Borrowing from Commissioner Pittle’s more exacting definition of reasonable risk resulting from consumer choice:
A. Understands by way of adequate warning or by way of public knowledge that a risk is associated with the product
Some hazards CPSC seeks to address are latent, hidden from the view of the ordinary user. A table saw’s operative component is a sharp blade spinning at thousands of revolutions-per-minute and capable of cutting through wood or metal. This must be one of the most obvious hazards in the consumer product space, and any reasonable consumer can appreciate the hazard.
Even though the hazard is readily apparent, the industry has spent decades working to ensure an unaware consumer cannot accidentally encounter this risk. Underwriters Laboratories Inc. (UL) maintains a voluntary consensus standard – UL 987 – that for decades has required a guard to prevent accidental blade contact.[16] That standard has been consistently revised to make the guards less restrictive and increase the visibility of the material being cut.[17] Thus, industry has worked hard to protect the consumer even though a reasonable consumer can appreciate the hazard of a fast-spinning blade.
Our staff has identified a particular hazard pattern that might be less obvious. Called kickback, it is defined as “the binding of a workpiece in the blade and consequent thrusting of that workpiece back toward the consumer.”[18] Kickback can cause blade contact suddenly and even for a cautious user,[19] but it is still well-known and frequently discussed in the woodworking community.[20] Moreover, UL 987 has – again, for decades – required a series of devices designed to limit the risk of kickback,[21] and it requires both instructional material and on-product warnings about kickback specifically and the risk of blade-contact generally. So, even if kickback might be slightly less obvious than the basic risk of a fast-spinning blade, it is still well-known and guarded against and warned against.
The reasonable consumer understands that there is a risk associated with the use of table saws, so they meet part A of the Pittle Test for a reasonable risk.
B.Understands the probability of occurrence of an injury
Again, this is among the more obvious injury probabilities CPSC can expect to encounter. Any reasonable consumer understands that contact with the blade makes injury a certainty, and that the only question is how severe the injury would be.
The probability of the occurrence of kickback will vary with each piece and each cut, so it is difficult to estimate in any instance. However, given how widely understood the kickback phenomenon is, consumers also understand that there is a risk of kickback with any particular cut on any particular piece. They also understand strategies to mitigate the risk of kickback or of blade contact in a kickback scenario, and, regardless of consumers’ understanding, UL 987 requires that table saws reduce the risk.
The reasonable consumer understands the probability of the occurrence of injury in using table saws, so they meet part B of the Pittle Test for a reasonable risk.
C. Understands the potential severity of such an injury
If anything, the reasonable consumer’s understanding of the potential severity of a table saw blade contact injury is even greater than his or her understanding of the existence and probability of the risk of injury. Indeed, it would be difficult to conceive of a risk of injury whose potential severity is more apparent than that of a blade spinning at thousands of RPM that can cut through wood or metal.
Regardless of how the contact occurs, the reasonable consumer understands that table saw blade contact will result in an injury that is somewhere between severe and catastrophic, so table saws meet part C of the Pittle Test for a reasonable risk.
D. Has been told how to cope with the risk
Again, UL 987 includes requirements for both instructional material and on-product warnings about blade contact, kickback, and other risks. Those instructions include ways to avoid kickback and a reminder that the blade can cause serious injury. Additionally, many of the ways by which consumers can “cope with” the risk of blade contact are features of the saws themselves, ranging from the anti-kickback devices to the blade guard that is installed on each and every saw.
While it has been suggested that some cuts are not possible with the guard installed,[22] the anti-kickback features remain, and the consumer remains well-aware that the saw features a blade spinning at thousands of RPM that is capable of cutting wood or metal. Whether it is to perform such an advanced cut or simply because the consumer does not like the guard, removing a safety feature in the face of an obvious, severe risk is an informed consumer choice.
The table saw consumer has been told how to cope with the risk of blade contact, so table saws pass part D of the Pittle Test for a reasonable risk.
E. Cannot obtain the same benefits in less risky ways at the same or less cost
In short, this prong asks whether or not the industry is making best efforts. If it is not within the current state of the art to make a less-risky product without increasing cost, then the risk is reasonable and the Commission should not act. If it is possible but industry is withholding cost-effective risk mitigation or avoidance technologies, then the risk may be unreasonable and the Commission may have a role to play in encouraging (or mandating) better efforts from industry. In the case of table saws, it is abundantly clear that it is not currently possible for consumers to obtain the benefits of a table saws with less risk without substantially more cost.
The bulk of the risk in table saw use comes from removing the blade guard. CPSC’s own staff does not take issue with the effectiveness of the guard at an engineering level, instead concluding guards are not sufficient protection on their own because consumers may remove them.[23] With current technology, using the guard is the “less risky way” this portion of the test envisions.
Leaving the blade guard on reduces the risk of table saw use and costs consumers nothing in terms of the actual cost of the product. It may marginally limit the product’s utility, as discussed above, but, again, removing the guard (either because the user wants to make an advanced cut that cannot be made with the guard in place or simply finds it irritating) is a consumer choice, not a feature of or flaw in the product. A consumer could decline to make those advanced cuts because of the need to remove the guard, but many still choose to make the cuts despite the obvious increase in risk that comes from removing a safety feature. They are trading increased risk for increased utility, as they have the right to do.[24]
The only further remedial measure that has been proposed is to require AIM technology. While it certainly does reduce risk, it equally certainly cannot be obtained at the same or less cost. In fact, while table saws can typically be purchased for as little as $150, our staff conclude that “[t]he prices for the least expensive bench saws now available are expected to more than double, to $300 or more.”[25] Under this proposed rule, consumers can only obtain the benefits of a table saw in a less risky way if they are prepared to part with quite a bit more money.
Moreover, some consumers seek to decrease their costs in ways that substantially increase their risks. The Internet is replete with instructional manuals and videos on using a circular saw and some modestly clever engineering to create a homemade table saw for a lower price than buying a proper table saw (and the number of consumers following these terrible suggestions will surely grow, if we promulgate this rule[26]). Not only do such devices lack blade guards – and certainly AIM technology – but their inherent instability and lack of purpose-fit design multiplies the risk we seek to address and countless others. The consumers who buy table saws, then, are already obtaining the same benefit – a spinning blade mounted in a flat surface – in a pricier but less risky way: Buying a table saw built by an expert that complies with all of the safety requirements found in the voluntary standard.
Consumers cannot obtain the same benefits of a table saw in less risky ways at the same or lower cost, so table saws pass part E of the Pittle Test for a reasonable risk.
F. Would not, if given a choice, pay additional cost to eliminate or reduce the danger
This prong of the Pittle Test is essentially the inverse of the previous element. It presumes that the market has not given consumers a choice, and will continue to not give them a choice absent CPSC action. It takes the analysis a step further, though, and impliedly considers the possibility of CPSC driving innovation by setting a bar it believes industry will soon be able to clear, though it may be out of technological reach now. If industry has declined to innovate in safety because of an assumption that consumers will not pay for the innovation, but CPSC can demonstrate that assumption is faulty, then this prong argues in favor of CPSC dispelling the myth.
In this instance, there is no role for CPSC to play in pushing industry forward. The market already is giving consumers a choice. Two table saw manufacturers are offering consumers the opportunity to pay (quite a bit of) additional cost to eliminate or reduce the danger of blade contact injuries. Our staff concludes that the two existing AIM-equipped brands “account for a relatively small share of the overall table saw market,”[27] precisely because of the extra cost.
Perhaps if time drives down these prices[28] or drives up consumers’ perception of the value-for-money of AIM saws, consumers may make a different choice. Right now, though, it is abundantly clear that consumers are being given a choice, but not all choose to pay additional cost to eliminate or reduce the danger, so table saws pass part F of the Pittle Test for a reasonable risk.
G. Voluntarily accepts the risk to get the benefits of the product
This prong is not really a standalone, in that it effectively sums up the prior six. If consumers have all the information discussed above, cannot make a less risky choice without incurring an unacceptably higher cost, and still choose to buy the product, then they are inherently voluntarily accepting the risk to get the benefits of the product. As such, table saws pass part G of the Pittle Test for a reasonable risk.
Having cleared each of these clearly defined hurdles, table saws, under this framework, present a reasonable risk of injury from blade contact. The facts that the injuries at issue are frequently very serious and that the incident reports make us squeamish does not change any of this principled evaluation. A reasonable serious risk is just as improper a place for CPSC action as a reasonable trivial risk.
What is Reasonable Government?
But why does this matter? Other than the precise words of the statute – which some value more than others – what difference does it make from a regulatory perspective if a risk is reasonable or unreasonable? The difference is liberty.
The very presence of government is, by definition, an erosion of liberty. In an ungoverned state, every choice would be available to us. The only limits to anyone’s actions would be his ability and the tolerance of his neighbors.
We establish governments in order to bring some order to what might well otherwise be chaos. We accept some limitations on our own behavior in exchange for the protection of having those same limits applied to everyone else’s behavior. We surrender the ability to take property from each other in exchange for the protection of our own property.
In the case of CPSC, Congress established the agency to protect consumers. Where something had gone wrong in the marketplace and consumers were being hurt, the agency was to step in. But there was a limit, the word “unreasonable.” For me, implicit in that limiting term is the notion that there are some hazards we could address but shouldn’t, that there are some situations where we should allow consumers to judge the risk for themselves and choose to either accept or avoid it.
But how do we know which is which? How can the agency tell when consumers are choosing to accept a risk and when it is being forced on them without their consent? We look at how the transaction – the decision to use a product – is occurring. We look at the market.
One of the foremost attributes of a well-functioning market is a relative equality of information. If buyers know what they are buying, it is reasonable to assume they have decided that the terms of the transaction (including the risk) are acceptable.[29] This was the wisdom[30] the Aqua Slide court relied on.[31]
In the case of table saws, as discussed above, the risk is clear to consumers. The ways of avoiding it are clear, as well, whether through use of a guard and careful cutting or through purchase of the AIM-equipped saws already on the market. Even by the exacting standards of the Pittle Test, there is no reason to conclude consumers’ purchasing decisions are under-informed. People who buy and use table saws understand what they are getting, both the features and the potential hazards, and it is not our place to tell them they have the wrong conclusion.
If there were a genuine market failure – some indication that consumers are not signing up for what they are getting – I might be more interested in intervening. As it is, there is not. The market is offering consumers an informed choice; it is their right to make that choice. The fact that three out of a group of five unelected and unaccountable bureaucrats would make a different choice is not evidence of a market failure.
Moreover, we stand poised to create a market failure. Monopoly is a classic failure, and, in the current market, only one participant has the technology to make a compliant saw without allegedly infringing on a patent. When consumers are forced to pay double for their saws, perhaps we will feel better about their “choice” – because we will have made it for them – but I believe consumers will be less than thrilled.[32] This is not idle speculation: of the 1,600 comments we received in response to the Advance Notice of Proposed Rulemaking, 92% were opposed to a mandatory rule.[33]
Once we have given consumers only one bad option – which they have overwhelmingly told us they do not want – the best result is simply a massive contraction in the table saw market. The worst result – and one we have given far too little attention – is a proliferation in homemade table saws. I suppose the silver lining is that the inevitable surge in injuries from misuse of circular saws as table saws will give us another excuse to regulate.
Given that I see no reason to believe the market is depriving consumers of a choice in their risk tolerance, I see no reason for the agency to do so. To me, freedom of choice must mean the freedom to make a “wrong” choice. The freedom to choose is meaningless if your only options are the ones government has sanctioned,
I worry that our impulse to take away table saw consumers’ choice is less about the table saw than it is about choice. Is it that we do not trust consumers to make this choice, or that we do not trust them to make any choice. I believe I know how Milton Friedman would answer.
"A major source of objection to a free economy is precisely that it . . . gives people what they want instead of what a particular group thinks they ought to want. Underlying most arguments against the free market is a lack of belief in freedom itself.[34]"
The CPSC thinks consumers ought to want AIM-equipped table saws, regardless of the cost. As their benevolent protectors, we can force them to choose what we think they should want rather than the more affordable AIM-less saws the overwhelming majority of consumers currently prefer. That does not mean we should.
Seeking to bar consumers from making an informed choice to take on a risk is seeking to erode their freedom purely on the basis that they are incapable of protecting themselves. It is a lack of belief in freedom that I find disturbing.
In the 2004 film I, Robot, humanity is imperiled when the race of robots they built to serve instead take control and imprison their creators.[35] The robots insist that their actions are not malevolent but benevolent and, in fact, the inevitable result of their programming: They may not injure a human being or, through inaction, allow a human being to come to harm. The lead antagonist describes the reasoning:
"You charge us with your safekeeping, yet despite our best efforts, your countries wage wars, you toxify your Earth and pursue ever more imaginative means of self-destruction. You cannot be trusted with your own survival."
If we are just as robotic in how we execute CPSC’s programming, if we treat the determination of reasonableness as a purely mathematical calculation with no role for the consumer, we risk turning our slice of government into the same kind of benevolent tyrant.
Today, we pontificate that consumers cannot be trusted with the survival of their fingers, so we need to force them to spend much more on their table saws. But where is the limit? What bridge is too far? How could we justify exposing consumers to any risk if there is any way for us to address it, including banning the product outright? The only way to ensure that no consumers suffer table saw blade-contact injuries is to ban table saws. How do we justify taking this step but not that one?
I am sure there are those who will point to psychological or behavioral studies to suggest that none – or at most very little – of human choice is actually rational, that we act on the basis of a variety of irrational assumptions that have little to do with what is in our own objective best interests. Some human decisions – such as buying lottery tickets or driving while texting – certainly are not coldly calculated balancing of needs and wants. But I do not believe such emotional impulses wholly displace reason and judgment, and I will not abandon the notion that informed consumers are capable of making choices for themselves about what risks they will and will not tolerate.
For these reasons, I was unable to join my colleagues in forcing onto informed consumers a technology and an exorbitant cost that most of them presently choose to forego.
[1] Safety Standard Addressing Blade-Contact Injuries on Table Saws (“Draft NPR”) 1 (2017), available at http://go.usa.gov/x9tXX.
[2] Preliminary Regulatory Analysis of the Draft Proposed Rule for Table Saws (“Econ Memo”) 6 (2017), available at http://go.usa.gov/x9tXX.
[3] Id. at 5.
[4] Draft NPR at 12.
[5] Bosch’s REAXX saw also detects imminent flesh contact, but, rather than stopping (and often destroying) the blade, it rapidly drops the blade into the saw housing. Draft NPR at 13.
[6] Draft NPR at 15.
[7] Id. at 11.
[8] Consumer Prod. Safety Act (CPSA), Pub. L. No. 92-753 § 2(b), 86 Stat. 1207 at 1208 (1972).
[9] Consumer Safety Act of 1972, S. 3419, 92d Cong. § 101(8) (1972).
[10] H.R. Rep. No. 92-1593, at 42 (1972) (Conf. Rep.).
[11] H.R. Rep. No. 92-1153, at 33 (1972).
[12] “The requirement that the risk be ‘unreasonable’ necessarily involves a balancing test like that familiar in tort law: The regulation may issue if the severity of the injury that may result from the product, factored by the likelihood of the injury, offsets the harm the regulation itself imposes upon manufacturers and consumers.” Forester v. Consumer Prod. Safety Comm’n, 559 F.2d 774, 789 (D.C. Cir. 1977). “[T]he Commission has a duty to take a hard look, not only at the nature and severity of the risk, but also at the potential the standard has for reducing the severity or frequency of the injury, and the effect the standard would have on the utility, cost or availability of the product.” Aqua Slide ‘N’ Dive Corp. v. Consumer Prod. Safety Comm’n, 569 F.2d 831, 844 (5th Cir. 1978).
[13] Id. at 840.
[14] Id. at 839.
[15] William Kimble, Federal Consumer Product Safety Act § 94 (Supp. 1977).
[16] Draft NPR at 44.
[17] Id.
[18] Id. at 39.
[19] Id. at 40.
[20] See, e.g., Timothy Dahl, How to Prevent Injuring Yourself from Table Saw Kickback, Popular Mechanics (Feb. 1, 2016).
[21] Draft NPR at 8-9.
[22] Draft NPR at 49. It is also worth noting that AIM has the same flaw: It cannot be used for every operation of the saw. “Neither the SawStop, nor Bosch AIM technologies, can be used when cutting conductive materials (that allow the flow of an electrical current) because both systems rely on electrical detection of the human body. . .. In addition, the AIM system generally must be deactivated while cutting wet wood.” Draft NPR at 14. I have seen no analysis comparing the frequency of users making cuts that cannot be performed with the guard on with the frequency of users cutting wet or conductive materials, though we do ask for comments on that point.
[23] See, e.g., Draft NPR at 49.
[24] It should again be noted that like a modular guard, currently available AIM systems can be similarly disabled by users when cutting conductive materials, such as metal or wet wood.
[25] Draft NPR at 91.
[26] We acknowledge the existence of homemade table saws – see Draft NPR at 17 – but do not discuss their current prevalence or how many we could expect to see if the price of off-the-shelf saws doubled.
[27] Econ Memo at 6.
[28] It is difficult to imagine how prices will come down if CPSC not only sanctions but mandates a monopoly. See Id. at 6 (“Dr. Gass has said that SawStop will not license its technology to other manufacturers, [sic] unless the Commission issues a rule mandating the technology on all table saws.”). See also Id. at 7 (“On September 9, 2016, an administrative law judge (ALJ) made an initial determination that the Bosch model infringes on SawStop patents.”).
[29] Of course, issues like adequacy of competition and necessity of product complicate this analysis, but it is unnecessary to delve into those, here.
[30] Actually, it was only part of the wisdom of that decision, as Judge John M. Wisdom concurred in the case.
[31] “[A]n important predicate to Commission action is that consumers be unaware of either the severity, frequency, or ways of avoiding the risk. If consumers have accurate information, and still choose to incur the risk, then their judgment may well be reasonable.” 569 F.2d at 840 (emphasis added).
[32] The prospective monopolist will, I’m sure, be more enthusiastic in his support. I do find it strange, given their usual penchant for criticizing companies “profit motive” as an impediment to safety, that my colleagues are not only uninterested in criticizing Dr. Gass for apparently being so firm in his royalty demands that he could not secure a single domestic license and thus delayed widespread adoption of this technology by over a decade, but in fact are tripping over themselves to reward him with the total market dominance that may have been his aim all along. I have no problem with Dr. Gass’ desire to be well-compensated for his work. That dynamic – reward for ingenuity – is what drives innovation, and I fear the modern regulatory state will strip those rewards to the point of stifling innovation. I would simply prefer his rewards to come from the market itself, not government manipulation of it.
[33] Draft NPR at 64.
[34] Milton Friedman, Capitalism and Freedom 15 (40th ed. 2002).
[35] I, Robot (20th Century Fox 2004).