(dissenting in part).
The social and economic setting has changed substantially since the collateral source rule first appeared in the middle of the last century. Its continued extensive use today has been criticized by most neutral commentators. See, e.g., Sedler, The Collateral Source Rule and Personal Injury Damages: The Irrelevant Principle and the Functional Approach, 58 Kentucky L. J. 36 (1969); Note, Unreason in the Law of Damages: The Collateral Source Rule, 11 Harvard L.Rev. 741 (1964); Fleming, The Collateral Source Rule and Loss Allocation in Tort Law, 54 Calif.L.Rev. 1478 (1966). Dean Richard C. Maxwell, in his article, The Collateral Source Rule in the American Law of Damages, 46 Minn.L.Rev. 669, 695 (1962), concludes, with some reservations, that the rule at best “operates as an instrument of what most of us would be willing to call justice.”
Our court has consistently applied the collateral source rule in a variety of situations; in my view, however, there is no *828good reason to extend the collateral source rule to this case. To the extent Dahlin v. Kron, 232 Minn. 312, 45 N.W.2d 833 (1950), suggests otherwise, it should not be followed.
Mr. Hueper has been compensated for his own personal injuries. He settled for $85,-000. Before us is his claim for hospital and medical expenses incurred by him for his son Bruce’s care during Bruce’s minority. This claim is separate from, although derivative of, Bruce’s claim. The jury awarded Mr. Hueper $43,847.34. This was reduced by 15% for Mr. Hueper’s own fault to $37,-370.24. This figure includes $24,977.14, representing the reasonable value of the services rendered by Shriners Hospital. Mr. Hueper keeps this money, since the hospital made no charge and accepts no payment for its services.' The $24,977.14 is to be paid by defendants Dean Goodrich and John M. Neubauer, the owner and driver, respectively, of the other truck involved in the accident. Mr. Goodrich is not personally at fault. He is liable to Mr. Hueper for the unincurred hospital bill of $24,977.14 by reason of his vicarious liability as owner of the truck.
The collateral source rule is usually justified on one or more of the following rationales: (1) the injured party has paid for the collateral source benefit and deserves what he paid for; (2) a collateral source benefit is sometimes a gift intended by the donor to benefit the injured donee and not the tort-feasor; (3) only cumulation of collateral source benefits with amounts assessed the tortfeasor will fully compensate the injured person; (4) a tortfeasor deserves to be punished, a purpose which would be foiled if he were relieved from “total responsibility” for his wrong; and (5) since a windfall payment is inevitable, better it go to the injured person than the tortfeasor.
None of these reasons, it seems to me, applies in this case, or, to the extent any does, it applies with very little persuasive force.1
First, Mr. Hueper did not pay for the Shriner’s care in any way. He paid no insurance premium; there was no surrender of sick leave or other fringe benefits. No “consideration” was given. Thus the case is unlike Van Tassel v. Horace Mann Ins. Co., 296 Minn. 181, 207 N.W.2d 348 (1973).
Second, in providing its care, the Shriners Hospital was indifferent as to whether Bruce Hueper had been in a compensable auto accident or not. He was afforded no different care than other patients. No specific gift was intended, such as where fellow employees donate blood to one of their number, as in Local 1140 v. Massachusetts Mutual Life Insurance Co., 282 Minn. 455, 165 N.W.2d 234 (1969), or where one spouse renders nursing care to the other.
Third, it cannot be said deduction of the Shriners Hospital bill will leave Mr. Hueper less than fully compensated. It is said that pain and suffering are more likely to be completely compensated if collateral source benefits are recoverable. Here, however, plaintiff’s entire claim is for specific expense items; no general damages, difficult to measure, are involved. The claim is more akin to a property damage claim where often the collateral source rule is disallowed.
Fourth, there would seem to be little reason to punish the defendants. If the defendants deserve punishment, punitive damages should be the route. If the jurors had known the hospital bill was not actually incurred, it is doubtful they would have *829punished defendants by making them pay it nevertheless. In any event, there seems to be little need to punish defendant Goodrich, who was not personally at fault.
Fifth, the windfall rationale lacks persuasive force here. Ordinarily, other things being equal, if someone is to benefit by the generosity of the Shriners Hospital, it is better that it be the injured party rather than the wrongdoer. What this means is that it is better to assure that plaintiff is made whole than that the defendant pay less than he ought to. Here, however, “other things” are not equal. Since plaintiff’s damages relate solely to “out-of-pocket” expenses, it cannot be said plaintiff is not made whole even with the collateral source benefit deducted. To say defendant should not pay less than he ought to is only another way of saying the defendant should be punished, a rationale which, as we have seen, is inappropriate in this factual setting.
While the collateral source rule has been strongly criticized, it is still true no one has really offered a better alternative. Perhaps this is the strongest rationale for its continued use. In most cases there is simply no way of knowing what is the fairest way to handle the collateral source benefit “windfall.” To try and find this “equality” by admitting a multitude of collateral sources would be most confusing, prejudicial and unfair.2 It is probably better, arbitrarily, to keep collateral source benefits out of the lawsuit and then allow the windfall to plaintiff.
But having said this, I see no need to extend this somewhat arbitrary rule to a fact situation where the policy considerations favoring it clearly do not apply. I would be willing to hold that in a parent’s derivative claim for out-of-pocket expenses the parent may not recover for hospital or medical care furnished his or her minor child gratuitously and not as a specific gift. At the very least, I would hold, in this case, that plaintiff is not entitled to recover against defendant Goodrich who is not personally at fault. It should be plaintiff’s option whether to prove up the reasonable value of the gratuitous services and then have the court, after the verdict, make the reduction, or to choose not to include the services in his claim.
. A collateral source is a benefit conferred by a third party. If plaintiff by his own efforts reduces or mitigates his damages, he cannot recover the reduction from the defendant. If plaintiff chooses to take a bed in a double room at the hospital rather than a private room, he cannot recover the difference from the defendant.
Appellants here argue that the collateral source rule applies to “injured persons” and that Mr. Hueper is not injured, not in the sense he sustained a personal injury. The argument is too technical. Mr. Hueper is “injured” in the tort sense of having incurred expenses for his son’s care, and this is sufficient. It might be argued Mr. Hueper has not “incurred” the hospital expense and so is not “injured,” but this is only another way of questioning the rationale behind the collateral source rule.
. If one starts recognizing collateral sources, where does one stop? Studies indicate that about 55% of auto accident tort victims’ compensation comes from tort recoveries, the remainder being supplied by collateral sources. Birmingham, The Theory of Economic Policy and the Law of Torts, 55 Minn.L.Rev. 1, 9 (1970); United States Department of Transportation, Automobile Insurance and Compensation Study, March 1971, “Motor Vehicle Crash Losses and their Compensation in the United States.” What about health insurance benefits, pensions, social security, sick leave, disability insurance and, in the case of a decedent, life insurance? And if these items are recognizable, then what about defendant’s liability insurance or lack of it? Here, for example, the Goodrich truck had only $100,000 coverage for any one claim and the insurer paid the $100,000 to Bruce, so apparently the defendants have personal exposure for Mr. Hueper’s verdict of $37,370.24.
“The statement of the problem as a conflict between overcompensation and windfall misses the mark if it assumes that absent this problem, there would be accurately-measured compensation. The most that can be done is to talk in terms of probability and risk. Do we increase the risk of overcompensation (or reduce the extent of undercompensation) by excluding consideration of benefits from a collateral source, or do we increase the risk of under-compensation (or reduce the extent of overcompensation) by allowing such benefits to be considered?” Sedler, The Collateral Source Pule and Personal Injury Damages: The Irrelevant Principle and the Functional Approach, 58 Kentucky L.J. 36, 59 (1969). The collateral source rule says, in effect, that since there is no ready answer to this question it is better not to ask the question.