¶1Dennis H. Shafer, argued, Shafer, Koch & Juidici, P.C., Negaunee, Mich., for defendant-appellant.
¶2A.H. Wisti, Hancock, Mich., Roger W. Zappa, argued, for plaintiffs-appellees.
¶3Before MERRITT and RYAN, Circuit Judges; and BROWN, Senior Circuit Judge.
¶5This products liability case applying Michigan law resulted in a jury verdict awarding the plaintiffs $1,200,000, reduced by 90% for contributory negligence, for injuries suffered to Mr. Siminski’s back when he fell while wearing a safety belt manufactured by the defendant Klein Tools, Inc. The issue on appeal is whether the District Court erred in denying defendant’s motions for a directed verdict and for judgment notwithstanding the verdict on the ground that plaintiffs failed to present sufficient proof of defect in the product to make out a prima facie case under Owens v. Allis-Chalmers Corp., 414 Mich. 413, 326 N.W.2d 372 (1982). We decide that plaintiff did not meet the standard set forth in Owens, and therefore reverse.
¶6While working as a carpenter, Mr. Simin-ski slipped and fell. His fall was broken by the Klein safety belt he was wearing, at which point plaintiff felt a “jolt” and experienced sharp back pain. Subsequently, he *357endured extensive and persistent pain that resulted in prolonged disability and an enforced change in job to a lower paying occupation. Mr. Siminski has endured prolonged pain and suffering and his wife sought compensation for loss of consortium. The theory of plaintiffs’ case was that the design of the Klein belt was too narrow in width, that a wider belt would have distributed the force of the broken fall more widely and thus avoided focuss-ing the shock forces of his broken fall into the area of his back that was injured. Defendant countered that the belt was available as a component that could be and normally was combined with a wider “comfort pad” that would cushion shock forces; the pad was available to Mr. Siminski at the work site and known to him.
¶7During pretrial discovery, defendant’s counsel experienced great difficulty in identifying and deposing the expert witness upon which plaintiff was to rely. The District Court had ordered that plaintiff identify his expert by May 19, 1986 and make him available for deposition by June 18. Defendant was placed on a similar schedule, with which it complied.
¶8Plaintiff originally identified an expert named Pucholski, but as the time for his deposition approached withdrew him. On June 30, another expert named Butler was identified. However, when on July 9 defense counsel traveled to Toledo to depose Mr. Butler, he was unprepared to answer questions.
¶9Before trial commenced about a month later, the District Court ruled on motion made by the defendant to bar plaintiffs from offering any expert testimony and, further, to dismiss the lawsuit. Judge Miles denied the motion to dismiss, but granted the first motion and costs, explaining later that “by hedging deadlines” and disregarding the “unambivalent scheduling orders” and deadlines set by the court, plaintiffs’ counsel had not “play[ed] the game fairly.” App. 96. Judge Miles noted that he considered the resultant “handicap in making their case” a “suitable ‘punishment’ ” for plaintiffs. Id.Defendant does not claim on appeal that the District Court abused its discretion in barring expert testimony as a discovery sanction.
¶10At trial, the only testimony on defect offered by the plaintiffs was that of Mr. Siminski:
Q: [W]hy didn’t you think [the belts] were any good?
A: Because I felt they were too narrow.
Q: If you had had a wider belt, what would have happened?
A: Well, my opinion is that — it wouldn’t have been so much force in one little spot. I am convinced in my own mind that if you had a wider belt, it would cushion it more, spread it out more.
Q: Do you believe that you would have been injured as seriously if you had had a wider belt?
A: I doubt it. I don’t think so.
¶11App. 133.
¶12The District Court, noting that the parties conceded that “this case does not present the same complicated issues of causation as are usually present in defective design cases,” decided that this testimony, coupled with other circumstantial evidence in the case that would be available to the jury, sufficed to constitute the prima facie case required by Owens v. Allis Chalmers. We disagree.
¶13In Owens v. Allis Chalmers, the Michigan Supreme Court required that a plaintiff’s prima facie case include
data or other factual evidence concerning the magnitude of the risks involved, the utility or relative safety of the proposed alternatives, or evidence otherwise concerning the “unreasonableness” of risks arising from [the alleged defect].
¶14414 Mich. at 431-32, 326 N.W.2d 372.
¶15The District Court concluded that this burden could be satisfied by (1) testimony about risks provided by one of defendant’s witnesses, and (2) a “plethora” of circumstantial evidence which the jury could “examine, ruminate, and then cultivate with their own common sense knowledge of human events and the workings of things,” including “an examination of the rebar assembly and the Klein Tools catalog which *358contained depictions of the belt and other safety assemblies.” We disagree.
¶16In Prentis v. Yale Manufacturing Co., 421 Mich. 670, 365 N.W.2d 176 (1984), the Michigan Supreme Court frankly adopted the “risk-utility” test proposed a couple of decades before by Prof. John Wade of Vanderbilt Law School. 365 N.W.2d at 185-86; see Wade, On the Nature of Strict Tort Liability for Products, 44 Miss.LJ. 825, 834-35 (1973).
¶17The Prentis court candidly described its decision as one to adopt a “pure negligence” test in design defect cases, one which would require the plaintiff “to pass the higher threshold of a fault test in order to threaten the entire product line.” 365 N.W.2d at 185-86.
¶18The Prentis court further noted that
[U]nlike manufacturing defects, design defects result from deliberate and doc-umentable decisions on the part of manufacturers, and plaintiffs should be able to learn the facts surrounding these decisions through liberalized modern discovery rules. Access to expert witnesses and technical data are available to aid plaintiffs in proving the manufacturer’s design decision was ill considered.
¶20We believe that Prentis and Owens, taken together, define a test for the prima facie case in Michigan design defect actions that plaintiffs simply did not meet.
¶21What was required was proof sufficient for a reasonable jury to balance the magnitude of the risk versus the feasibility of other design alternatives, or otherwise to weigh the “unreasonableness” of risks arising from the decision to manufacture the 1% inch belt as a component to be “mixed and matched” with a wider comfort pad as opposed to making only wider belts or to always fixing the comfort pad to the narrow belt. In order to go to the jury, some proof was necessary to show that making a narrow belt constitutes negligence. Instead the proof showed that many users want the narrow belt so that they can either use the manufacturer’s various components or components which they fabricate themselves. The only evidence of this type before the jury was that offered by Klein of this market demand for the narrow belt as a single component, and of complaints from workmen when this option was not available.
¶22We conclude that evidence of the type demanded by Owens and Prentis could have been obtained by plaintiffs in discovery; failure to offer it was fatal to their chance to go to the jury. Plaintiff’s bare opinion, coupled with mere examination of the catalog choices and belt was insufficient. The judgment of the District Court is reversed.