Kelly, J.
This appeal is a consolidation of two cases. In each, the primary issue is whether the Supreme Court should overrule or modify its decision in Fiser v City of Ann Arbor, 417 Mich 461; 339 NW2d 413 (1983).
The cases involve automobile chases in which the pursued vehicles, operated by individuals fleeing police officers, crashed into vehicles occupied by innocent parties. Under Fiser, if the police pursuit constitutes negligent operation of the police vehicle, then the motor vehicle exception applies1 and the municipality is not immune from tort liability. Fiser, supra at 469.
The motor vehicle exception to governmental immunity allows a governmental agency to be held [129]*129liable for its employee’s negligent operation of a government-owned vehicle. MCL 691.1405; MSA 3.996(105) states in pertinent part:
Governmental agencies shall be liable for bodily injury and property damage resulting from the negligent operation by any officer, agent, or employee of the governmental agency, of a motor vehicle of which the governmental agency is owner ....
In Ewing v Detroit (On Remand),2 defendants appeal from a decision of the Court of Appeals that reversed the trial court’s grant of summary disposition in their favor. The Court of Appeals held that plaintiffs in that personal injury case had presented sufficient evidence to raise genuine issues of fact regarding negligence and proximate causation.
However, defendants contend that the conduct of the police officers cannot be a proximate cause of plaintiffs’ injuries. They argue that the Court of Appeals erred in reversing the trial court’s decision, because plaintiffs had failed to present a prima facie case of negligence. We disagree with defendants.
Rather, we adopt the reasoning outlined by Justice Ryan in his concurrence in Fiser at 479-480. Justice Ryan stated, “It is for the factfinder to determine whether the actions of the operators of the pursuing vehicles were causes in fact of the plaintiff’s injuries.” Id. at 479. Thus, we hold that the high-speed pursuit of the McGuigan vehicle by the police may have proximately caused or proximately contributed to the injuries plaintiffs allege.
[130]*130The appeal in Rogers arises out of a wrongful death suit in which the jury awarded $6,132,993 in favor of the widow of John Rogers.3 Defendants argue that they can be held hable for no more than the officers’ negligent operation of their vehicle, not for their decision to commence or continue pursuit. They claim that the Court of Appeals erred in affirming the trial court’s decision that allowed plaintiff to proceed to trial on theories of liability because they are barred by governmental immunity. We disagree.
We find that the trial court did not err in admitting evidence pertaining to defendants’ police-pursuit policies. Under Fiser, a police officer must exercise reasonable care4 and may exceed the speed limit only “so long as he does not endanger life or property.”5 Id. at 471. Accordingly, we find that the trial court correctly denied defendants’ motion for directed verdict.
In Rogers, we granted leave to appeal, not only on the immunity issue, but also on issues concerning the verdict itself and various errors alleged to have occurred at trial. We also granted plaintiff’s cross-application on the issues of apportionment of fault and collateral source setoffs. We affirm the decision of the Court of Appeals regarding the misconduct issues, but reverse it concerning the apportionment of fault and collateral source setoffs.
We recognize that innocent people are endangered by police pursuits of fleeing vehicles. Also the Legisla[131]*131ture has not taken steps to statutorily overturn our decision in Fiser. Consequently, we adhere to the principle of stare decisis and uphold the rule as articulated in that case.
FACTS
EWING v DETROIT
On November 29, 1990, thirty-seven-year-old Deborah Ewing was driving a van through her residential neighborhood in northwest Detroit. Her four-year old daughter, Krystal, was with her. Ms. Ewing was eastbound on Florence going twenty to twenty-five miles an hour. She slowed as she approached the intersection of Florence and Shaftsbury, an “open” residential intersection three blocks south of McNichoIs.
As she entered the intersection, a 1985 Dodge Ram pickup truck broadsided her van. The truck had been traveling southbound on Shaftsbury at a high rate of speed, variously estimated at sixty to seventy and ninety to one hundred miles an hour. It was driven by Jay McGuigan who, at the time, was fleeing police officers.
As a result of the collision, both vehicles were rendered nondriveable. Mr. McGuigan fled from his vehicle on foot, but was chased, apprehended, and arrested by police officers. Ms. Ewing was rendered unconscious. Both she and her daughter suffered severe injuries.6
[132]*132The facts surrounding the police chase are as follows: Officer Michael Malott worked in the Detroit Police ministation section as part of a backup arrest unit for a surveillance crew. He was alone in a car designated as the “scout car.” The car was without police markings, but was equipped with red and blue grille lights, rear deck lights, and a siren. Several other surveillance units were nearby. They were manned by Officers Dettore, Robbins, and White. The units were conducting a surveillance of a known stolen pickup truck parked in a driveway on Margareta in a residential neighborhood.
The cars and their backups were in closed-channel radio communication with one another. When Officer White radioed to the others that the pickup truck was occupied and moving, they replied that they were going to stop the stolen vehicle. Officer Malott began to drive to the location to assist in the arrest. However, before he arrived, he learned that the driver of the pickup truck had escaped7 and was being chased by another unit.
A high-speed chase ensued down Shaftsbury through single-family residential neighborhoods. The pursued vehicle and, at times, the chasing police vehicle disregarded traffic signs and signals. Eventually, the pursuit ended with the collision of the pickup truck and the van driven by Ms. Ewing. Testimony [133]*133established that the police vehicle was approximately one-half block away at the time of collision.
Ms. Ewing brought suit against the defendant, city of Detroit, and its police officers, pleading negligent operation of government vehicles and gross negligence on the part of the individual police officers. Defendants filed a third-party complaint against Mr. McGuigan.
The city moved for summary disposition against Ms. Ewing, alleging governmental immunity and failure by plaintiffs to state an actionable claim and a dispute of fact submittable to a jury. The trial court granted summary disposition in favor of the city on both counts.
Plaintiffs appealed to the Court of Appeals. The panel8 peremptorily reversed the summary disposition in favor of the city and its police department, but affirmed summary disposition in favor of the individual police officers.
The city then appealed to this Court. We issued a remand order on May 2, 1995. 448 Mich 928. It stated:
In lieu of granting leave to appeal, the portion of the July 6, 1994, order of the Court of Appeals reversing the grant of summary disposition to the defendants city of Detroit and Detroit Police Department is vacated, and the case is remanded to that Court for plenary consideration only of the portion of plaintiff’s appeal from the Wayne Circuit Court’s order granting summary disposition to those defendants on plaintiff’s theory under MCL 691.1405; MSA 3.996(105).
On remand, the Court of Appeals again reversed the trial court’s grant of summary disposition to the [134]*134city and its police department. Defendants appealed to this Court and we granted defendants’ application for leave to appeal.9
ROGERS v DETROIT
This appeal arises out of a wrongful death suit in which the jury awarded $6,132,993 in favor of plaintiff. On June 17, 1988, John Rogers10 was driving his automobile west on Jefferson approaching Lemay. He was fatally injured when another vehicle that had been speeding on Lemay struck his vehicle. The driver of the other vehicle was fourteen-year-old Keith Montgomery, an unlicensed driver who was attempting to flee from Detroit police officers chasing him in their car. Mr. Montgomery allegedly had failed to stop at the intersection of Lemay and Kercheval.
Plaintiff brought suit against the city, its police department, and the police officers who undertook the pursuit. She claimed that the pursuit constituted a hazard and that the city and its police officers were liable for the death of her husband because of their negligence. They included the commencement and continuation of the pursuit, inadequate selection and training of officers, and failure to promulgate proper policies regarding vehicular pursuits. The allegations also stated that Mr. Montgomery was liable because of his negligent operation of the car. Plaintiff did not pursue the suit against him, presumably because he [135]*135was an uninsured minor and uncollectible at the time of the accident.
Pursuant to MCR 2.206,11 the city filed a motion to join Montgomery as a party-defendant. They asked the circuit court to retain him as a defendant by requiring plaintiff to serve him with a summons and copy of the complaint.
At a settlement conference, the circuit court stated that it could not force plaintiff to proceed against Mr. Montgomery. It allowed plaintiff to dismiss him as a defendant, thereby denying the city’s motion. However, the court allowed Mr. Montgomery to be added as a third-party defendant, because it recognized the city’s desire and right to secure a statutory apportionment of damages. Mr. Montgomery was then added as a third-party defendant only for the purpose of apportioning the city’s liability. MCL 600.6304; MSA 27A.6304.
The trial commenced on April 6, 1992, and ended on April 15, 1992, with the jury returning a verdict12 in favor of plaintiff. It found that defendants had been negligent and that the negligent operation of the police car was the cause of John Rogers’ death. Damages were assessed in the amount of $6,132,993. The jury apportioned fault, eighty percent to the city and twenty percent to Montgomery. Because Montgomery was uncollectible, the circuit court held that the city was responsible for payment to plaintiff of the entire verdict.
[136]*136The city then requested a new trial. The court denied the motion, but granted an evidentiary hearing on the question whether plaintiff and the third-party defendant had cooperated at trial.
After the hearing, the court denied the remainder of the city’s motion. It heard plaintiff’s later motion for entry of judgment and allowed plaintiff’s counsel ten days from the entry of judgment to notify contractual lienholders of the verdict. The order was notwithstanding statutory language requiring plaintiffs counsel to notify lienholders within ten days of the verdict. MCL 600.6303(3); MSA 27A.6303(3).
Another hearing was held at which defendant sought a statutory collateral source setoff, requesting that the judgment be reduced by $1,050,440. The court refused. Instead, it ordered that the disputed amount be placed in escrow when the judgment was paid. The court entered judgment without reducing the city’s liability by the twenty percent for which Montgomery was liable and without giving defendant the benefit of the collateral source setoffs. The city appealed to the Court of Appeals.
On appeal, the Court of Appeals affirmed the judgment, but ordered a remand with the direction that the court apportion the judgment pursuant to the verdict and allow collateral source setoffs. Plaintiff now appeals from that decision.
In addition to defendants’ Fiser claim, defendants also appeal from the Court of Appeals decision that refused to grant a new trial on the basis of various alleged errors perpetuated at trial. We granted defend[137]*137ants’ application for leave to appeal and plaintiff’s application for leave to appeal as cross-appellant.13
FISER AND ITS PROGENY
In Fiser, supra, this Court addressed what immunity is available to a municipality and its police officers from liability for injuries related to an automotive police chase. Specifically at issue was the interpretation and application of the “motor vehicle exception” to immunity. MCL 691.1405; MSA 3.996(105). In the years following Fiser, its analysis has dominated and controlled questions of immunity arising from police pursuits.
Fiser arose from a high-speed police chase through a residential neighborhood that ended when the fleeing suspect struck the plaintiff’s vehicle. This Court unanimously reversed a summary disposition order, holding that the plaintiff’s liability claims fell within the motor vehicle exception. MCL 691.1405; MSA 3.996(105).
In each of the cases before us, plaintiffs must prove that (1) defendants owed a duty to plaintiffs, (2) defendants breached that duty, (3) plaintiffs suffered damages, and (4) defendants’ breach of duty or negligence was a proximate cause of plaintiffs’ damages. Fiser, supra at 469-470.
Defendants have a duty to exercise toward plaintiffs that care that a reasonably prudent person would exercise while discharging official duties of a similar nature under similar circumstances. Fiser, supra at 470, quoting McKay v Hargis, 351 Mich 409, 418; 88 NW2d 456 (1958). When applying that standard, a [138]*138court considers the statutes governing operation of emergency vehicles. MCL 257.603; MSA 9.2303,14 MCL 257.632; MSA 9.2332.
Those statutes provide:
(b) The driver of an authorized emergency vehicle when responding to an emergency call, but not while returning from an emergency call, may exercise the privileges set forth in this section, subject to the conditions of this section.
(c) The driver of an authorized emergency vehicle may:
(1) Park or stand, irrespective of the provisions of this act.
(2) Proceed past a red or stop signal or stop sign, but only after slowing down as may be necessary for safe operation.
(3) Exceed the prima facie speed limits so long as he does not endanger life or property.
(4) Disregard regulations governing direction of movement or turning in specified direction.
(d) The exemptions granted in this section to an authorized emergency vehicle shall apply only when the driver of the vehicle while in motion sounds an audible signal by bell, siren, air horn, or exhaust whistle as may be reasonably necessary except as provided in subsection (e), and when the vehicle is equipped with at least 1 lighted lamp displaying a flashing, oscillating, or rotating red or blue light .... [MCL 257.603; MSA 9.2303 (emphasis added).]
The speed limitation set forth in this chapter shall not apply to vehicles when operated with due regard for safety under the direction of the police when traveling in emergencies or in the chase or apprehension of violators of the law or of persons charged with or suspected of a violation, nor to fire department or fire patrol vehicles when traveling in response to a fire alarm, nor to public or private ambu[139]*139lances when traveling in emergencies. . . . This exemption shall not however protect the driver of the vehicle from the consequences of a reckless disregard of the safety of others. [MCL 257.632; MSA 9.2332 (emphasis added).]
Subsections 603(c)(2) and (3) allow a driver of an emergency vehicle to proceed past stop signals and signs after slowing “as may be necessary for safe operation.” The driver may exceed the speed limit “so long as he does not endanger life or property.” According to Fiser, “[t]he legislative intent is expressed in these statutes — emergency vehicles must be driven with due regard for the safety of others.” Id. at 472. We infer from this language that a duty to innocent bystanders, onlookers, and travelers does exist.
The Fiser Court stressed that abstract policy was not to be considered in cases of this type because the liability in question is statutory in nature. It stated:
The defendants argue that a reversal of the decision of the Court of Appeals would be contrary to good public policy because it would have a chilling effect on the performance of police officers. Whatever the merits of defendants’ argument, it is not for us to decide. The Legislature has determined the public policy embraced in MCL 257.603, 257.632, 691.1405; MSA 9.2303, 9.2332, 3.996(105), and, if the performance of police officers is chilled thereby, undoubtedly the Legislature was satisfied that a higher public good was served by enacting the statutes. Even if we were persuaded that the policy is unwise, we would not be free to change it. [Id. at 475-476.]
Fiser's deference to the Legislature did not go unnoticed by lobbyists for municipalities. In the wake of Fiser, House Bill 4003 was proposed. It would have amended § 5 of 1964 PA 170 and was described as:
[140]*140An act to make uniform the liability of municipal corporations, political subdivisions, and the state, its agencies and departments, when engaged in the exercise or discharge of a governmental function, for injuries to property and persons; to define and limit this liability; to define and limit the liability of the state when engaged in a proprietary function; to authorize the purchase of liability insurance to protect against loss arising out of this liability; to provide for defending certain claims made against public officers and paying damages sought or awarded against them; to provide for the legal defense of public officers and employees; to provide for reimbursement of public officers and employees for certain legal expenses; and to repeal certain acts and parts of acts, being section 691.1405 of the Michigan Compiled Laws.
The Legislature rejected House Bill 4003, and, in the 1986 revisions of the governmental immunity statute, retained the language of MCL 691.1405; MSA 3.996(105) that was interpreted in Fiser.
Against this backdrop, we reject defendants’ request to overrule Fiser. Here, the statutory meaning is clear on its face. Therefore, the role of the judiciary is not to articulate its view of “policy,” but to apply the statute in accord with its plain language. Lorencz v Ford Motor Co, 439 Mich 370, 376; 483 NW2d 844 (1992). Also, where legislation has been authoritatively construed by the courts, then retained by the Legislature, we will find legislative concurrence. Margreta v Ambassador Steel Co, 380 Mich 513, 520; 158 NW2d 473 (1968).
Defendants argue that Fiser was overruled sub silentio by Ross v Consumers Power Co (On Rehearing), 420 Mich 567; 363 NW2d 641 (1984). We disagree.
The contention that Fiser was overruled misapprehends the meaning of Ross. The “discretion[141]*141ary/rninisterial” test of Ross applied only to the common-law liability of individual employees, not to the statutory liability of governmental agencies. Ross held that neither that decision nor MCL 691.1407; MSA 3.996(107) detracts from a city’s liability under MCL 691.1405; MSA 3.996(105), as applied to police pursuits in Fiser. Ross, supra at 591,15 593-594,16 620-622.17
We find that Ross could not and did not overrule Fiser. Thus, we conclude that Fiser remains good law in Michigan, and that the proofs under each of the cases at hand establish a cause of action submittable to a jury.
Our decision in Fiser expressly states that claims of negligence and proximate cause are to be submitted to and decided by a trier of fact and not by summary disposition. Id. at 471-476. Defendants, having failed to distinguish their cases from Fiser, have not persuaded us to overrule our decision in Fiser.
We further decline to narrow our findings in Fiser and believe that any limitation in liability must come from the Legislature.
[142]*142APPLICATION OP LAW
EWING v DETROIT
In Ewing, defendants are emboldened in their proximate causation analysis by the Court of Appeals criticism leveled against Fiser in Ewing and Frohman v Detroit,18 181 Mich App 400; 450 NW2d 59 (1989). Both the Ewing and the Frohman Court of Appeals panels expressly followed Fiser, but each criticized Fiser’s proximate causation analysis. In neither was the criticism unanimous. We are not persuaded by the criticism to alter our Fiser analysis.
Defendants attempt to draw support for their argument by relying on three other Michigan Court of Appeals police-pursuit decisions: Cooper v Wade, 218 Mich App 649; 554 NW2d 919 (1996); Robinson v Detroit (On Rehearing), 225 Mich App 14; 571 NW2d 34 (1997); Jackson v Oliver, 204 Mich App 122; 514 NW2d 195 (1994). However, we find these cases distinguishable from the cases before us.
Fiser, Frohman, and Ewing all involve tort actions brought by injured innocent bystanders during police pursuits, Jackson, Cooper, and Robinson involve actions brought by the operators or passengers of pursued vehicles. The latter cases dealt not with proximate causation, but with the issue of duty.
We reject the conclusion that, under the tragic facts of Ewing, it was, as a matter of law, the felon’s flight not the officers’ pursuit that proximately caused plaintiffs’ injuries. In Turner v Auto Club Ins Ass’n,19 [143]*143this Court addressed a similar question in a no-fault property protection insurance case.
Turner involved a police chase accident. As in Ewing, there was no physical contact between the pursuing police car and any of the colliding vehicles. We reversed a Court of Appeals decision that held that the accident resulted not from police use of a vehicle, but from the thief’s act of fleeing the police. Turner, supra at 42-43. We stated that “the use of the police vehicle as a motor vehicle had an active link with the damage, making it ‘involved in the accident’. . . .” Id. at 43. Thus, the police car was partly responsible. Id.
Therefore, we find Turner consistent with Fiser and conclude that there is nothing inimical about Fiser's analysis and the application of established proximate causation principles to police-pursuit cases.
There may be more than one proximate cause of an injury. Hence, if the facts are as alleged, the negligent conduct of Mr. McGuigan, and the negligent conduct of the police officers both were proximate causes of plaintiffs’ injuries. That Mr. McGuigan may have been an intervening agent would not absolve the city and its police department from liability. As stated in 2 Restatement Torts, 2d, § 447, p 478, and in Fiser, supra at 479-480:
The fact that an intervening act of a third person is negligent in itself or is done in a negligent manner does not make it a superseding cause of harm to another which the actor’s negligent conduct is a substantial factor in bringing about, if
(a) the actor at the time of his negligent conduct should have realized that a third person might so act, or
[144]*144(b) a reasonable maxi knowing the situation existing when the act of the third person was done would not regard it as highly extraordinary that the third person had so acted, or
(c) the intervening act is a normal consequence of a situation created by the actor’s conduct and the manner in which it is done is not extraordinarily negligent.
Thus, Fiser remains consistent with the Restatement and its established proximate causation principles. Moreover, ours is among the growing majority of jurisdictions recognizing that police pursuits can be the proximate cause of ipjuries to innocents and that proximate causation is a question for the jury. See, e.g., Carl v City of Overland Park, Kansas, 65 F3d 866, 875 (CA 10, 1995); Haynes v Hamilton Co, 883 SW2d 606, 612 (Tenn, 1994).
The Fiser Court has articulated several factors to be considered when evaluating the reasonableness of a police officer’s conduct in a police-pursuit situation. They are: (1) whether there was an emergency, (2) the speed of the pursuit, (3) the area of the pursuit, (4) the weather and road conditions, (5) the presence of pedestrians and other traffic, (6) the presence or absence of audible and visible warnings, and (7) the reason the officers were pursuing the fleeing vehicle. Id. at 472.
Taking these factors into account, we found in Fiser issues for the jury of negligence and proximate causation where the police-pursuit situation involved: (1) high speed, (2) disregard of traffic signals, (3) pursuit through residential and business areas, and (4) a wet pavement at a well-traveled intersection. Id. at 472-473.
[145]*145In the present case, the situation is similar. A high-speed chase took place: (1) across busy major city streets, (2) through densely populated urban residential areas, (3) during a time of day when schools released children, (4) in disregard of traffic and traffic signals, (5) by not fully marked police vehicles, (6) in violation of internal police standards, and (7) in pursuit of a stolen vehicle that the police had been surveilling.
Plaintiffs have pleaded facts in Ewing sufficient to state an actionable claim, including facts in avoidance of immunity. There exists a genuine issue of material fact concerning whether the officers were negligent and, if negligent, proximately caused plaintiffs’ injuries. Therefore, we find that the Court of Appeals correctly reversed the trial court’s grant of summary disposition for defendants. We affirm the decision and remand Ewing for trial.
ROGERS v DETROIT
GOVERNMENTAL IMMUNITY
In Rogers, defendants contend that the trial court erred by allowing plaintiff to proceed to trial on theories of liability that were barred by governmental immunity. The trial court allowed plaintiff to challenge the police officers’ familiarity and compliance with the department’s policies regarding chases. It also permitted jury consideration of whether the officers were negligent when they decided to continue the pursuit of Mr. Montgomery.20 Defendants [146]*146claim that governmental immunity applied to all acts of their officers, except those stemming from the negligent operation of a motor vehicle. They argue that they could not be held liable for the officers’ decision to commence or continue the pursuit. We disagree with defendants.
After Fiser, it is clear that, if a police pursuit constitutes negligent vehicular operation, then the motor vehicle exception applies and the municipality is not immune from tort liability. Fiser, supra at 469. Therefore, the jury’s finding of negligence on the part of defendants was reasonable if based on (1) the unduly high chase speeds reached by the police vehicle, (2) on congested residential and commercial city streets, and (3) when the officers failed to operate their siren and overhead lights.
We find it appropriate that the jurors were instructed to consider the police department policy, the officers’ knowledge of the policy, the applicable statutes, and their own everyday experience. The department policy was relevant. Defendants did not object to its admission. It was therefore properly admitted and properly used.
Accordingly, we find that the trial court correctly denied defendants’ motion for a directed verdict.
ALLEGED ATTORNEY MISCONDUCT
Defendants claim that repeated and extensive misconduct by plaintiff’s counsel during closing argument diverted the jury from the primary issues of the case and denied defendants a fair trial. We disagree.
This Court stated in Reetz v Kinsman Marine Transit Co, 416 Mich 97, 102-103; 330 NW2d 638 (1982):
[147]*147When reviewing an appeal asserting improper conduct of an attorney, the appellate court should first determine whether or not the claimed error was in fact error and, if so, whether it was harmless. If the claimed error was not harmless, the court must then ask if the error was properly preserved by objection and request for instruction or motion for mistrial. If the error is so preserved, then there is a right to appellate review; if not, the court must still make one further inquiry. It must decide whether a new trial should nevertheless be ordered because what occurred may have caused the result or played too large a part and may have denied a party a fair trial. If the court cannot say that the result was not affected, then a new trial may be granted. Tainted verdicts need not be allowed to stand simply because a lawyer or judge or both failed to protect the interests of the prejudiced party by timely action.
The first asserted improper conduct by plaintiff’s counsel is his characterization of defense counsel to the jury as “slick,” a “slick talker,” “smooth,” and “smoothtalking.” Plaintiff responds that all these comments were responses to argument by defendants.
Plaintiff’s counsel may have been attempting to persuade the jury that defense counsel was not being truthful in some of his arguments. However, we disagree that this constituted trial misconduct. Moreover, defendants did not object to the comments.
We agree with the Court of Appeals that the verdict was not changed by plaintiff’s counsel’s references to defense counsel. Plaintiff’s case was very strong. Counsel presented convincing evidence that the two police officers undertook a police chase that was improper and negligent. The jury believed plaintiff’s counsel and, we conclude, would have ruled for plaintiff as it did, irrespective of counsel's disparaging remarks.
[148]*148At the same time, we discourage any attorney from denigrating another before a jury. Such behavior jeopardizes the integrity of a trial and reflects adversely on the legal profession. Language unbecoming a court officer imperils the decorum that courts must maintain.
The second challenged error is plaintiffs counsel’s reference to facts not in evidence. During rebuttal argument, plaintiffs counsel read from a portion of Mr. Montgomery’s deposition not in evidence to rebut the claim that Montgomery had changed his testimony.
Although it was improper for counsel to read the testimony, defendants again failed to object. If defense counsel had objected, then the trial court could have immediately instructed the jury. On the basis of defendants’ failure to object and a lack of manifest injustice, we decline to reverse the trial court’s decision not to set aside the jury verdict.
Defendants object also to plaintiff’s counsel's placement on one of plaintiff’s exhibits of the letter B. It appeared on a large street chart showing the area traversed by the police chase.
During closing argument, defense counsel questioned the meaning of the B. During rebuttal, plaintiff’s counsel responded that he had placed the letter on the diagram to designate the point at which the police could have stopped their chase. Defendants objected.
The assertion that the B stood for the point at which the police should have discontinued their chase was first raised during closing argument. Plaintiff asserts that, because plaintiff’s counsel prepared [149]*149the exhibit, he was entitled to explain the placement of the letter. We agree.
Defense counsel made the marking on the chart a prominent theme of his closing argument. Thus, plaintiffs counsel’s explanation of it should not be viewed as evidence. Moreover, the judge instructed the jury that the attorneys’ arguments were not to be considered evidence. Therefore, we find no error.
Next, defendants assert that plaintiff’s counsel made comments during his closing and rebuttal arguments that inappropriately appealed to the sympathy of the jury. Whereas an appeal to the jury’s sympathy may constitute grounds for reversal, the resultant prejudice can be cured by an instruction. The jury can be advised that its verdict must not be influenced by sympathy. Jakubiec v Hasty, 337 Mich 205; 59 NW2d 385 (1953).
In Rogers, defendants neither objected to plaintiffs counsel's remarks nor requested an instruction. Moreover, the trial judge instructed the jury that “[s]ympathy must not influence your decision.” Thus, we decline to reverse on this ground.
The conduct of plaintiff’s counsel that defendant complains of in Rogers does not warrant a new trial. We find that it was the proofs, not the challenged argument, that led the jury to find the city liable.
THE ALLEGED RELATIONSHIP BETWEEN MR. MONTGOMERY AND PLAINTIFF
When plaintiff filed suit, he named Mr. Montgomery as a defendant. However, Montgomery was never served, and plaintiff did not proceed against him as a first-party defendant.
[150]*150Defense counsel then sought leave to file a third-party complaint against Montgomery. Plaintiff expressed concern that Montgomery was appearing without counsel and referred him to attorney Sheldon Miller.21
During trial, defense counsel raised questions about the relationship between plaintiffs counsel and Montgomery’s counsel. The trial judge stated that he would allow the defense to question Montgomery about his relationship with plaintiff and plaintiff’s counsel. However, for whatever reason, the defense chose to pose no questions.
Defendants now contend that they were denied a fair trial because counsel for plaintiff and Montgomery entered into a “Mary Carter”22 agreement. In Smith v Childs,23 the Court of Appeals stated:
The distinguishing characteristics of a Mary Carter agreement are that it (1) not act as a release, so the agreeing defendant remains in the case, (2) is structured in a way that it caps the agreeing defendant’s potential liability and gives that defendant an incentive to assist the plaintiff’s case against the other defendants, and (3) is kept secret from the other parties and the trier of fact, causing all to misunderstand the agreeing defendant’s motives.
[151]*151Mary Carter agreements deny a fair trial to those defendants who are not part of the agreement. Smith v Childs, supra at 97-98.
While the record may raise questions regarding the relationship between plaintiff’s counsel and Montgomery’s counsel, there is no direct evidence that a Mary Carter agreement was ever established. Defendants have failed to demonstrate that an agreement existed that capped Montgomery’s liability as an incentive for him to assist plaintiff in her case against defendants. Thus, the argument that such an agreement denied defendant a fair trial fails.
EVIDENTIARY RULINGS
Defendants also argue that the trial court erred in admitting several items of evidence denying them a fair trial. We find that the trial court properly admitted the evidence in question, including the photographs, videotape, and deposition testimony of a party.
After the trial had commenced, one of plaintiff’s expert witnesses visited the scene of the accident, took photographs, and shot a videotape. These items were introduced into evidence over the objections of defense counsel. The photographs and the videotape were relevant. They showed the jurors the geographical layout of the area, the length and width of the street, and the configuration of the intersection. Plaintiff did not purport, as defendants argue, that this evidence was a representation of how the streets appeared when the incident occurred, merely their layout. Therefore, defendants’ argument is misplaced.
[152]*152We find that Stuive v Pere Marquette R Co,24 is on point. In Stuive, the Court held that, when it is admitted that relevant evidence did not show conditions on the date of the accident, there is no error in its admission. Id. at 154. Therefore, we find that the photographs and videotape were properly admitted.
Regarding the deposition testimony of Officer Jackson, the trial judge allowed plaintiff to read into evidence certain portions of his testimony. The ruling was not erroneous because the deposition testimony constituted an admission by a party defendant. Thus, it is not barred by MRE 801(d)(2).
APPORTIONMENT OF THE VERDICT
A. WEIGHT OF EVIDENCE
Defendants argue that the finding of the jury that the city was eighty percent and Montgomery was twenty percent at fault was against the great weight of the evidence. We disagree.
To make its decision, the jury must weigh the testimony of the witnesses and determine their credibility. Johnson v Corbet, 423 Mich 304, 314; 377 NW2d 713 (1985).
The testimony at trial was that Montgomery’s initial offense was minor; he had squealed his tires before making a rolling stop at a stop sign. Both pursuing police officers testified that they did not learn that the Montgomery automobile had been stolen until after the pursuit terminated. In addition, the testimony showed that the police officers were supposed to, but failed to, activate their lights and siren when engaging in the automobile pursuit. The purpose of [153]*153the requirement is to warn other motorists, such as Rogers, of the chase. When Montgomery accelerated in a residential area during rush-hour, the officers sped up to continue the chase rather than discontinuing it.
The jury was entitled to conclude from that evidence that the city was eighty percent at fault. Therefore, we affirm the Court of Appeals in this regard.
B. AMOUNT OF DAMAGES
Plaintiff Rogers alleges that the Court of Appeals misconstrued MCL 600.6304; MSA 27A.630425 when it limited the city’s liability to paying no more than eighty percent of the verdict. Montgomery was uncollectible. Consequently, plaintiff argues, the statute requires that the city be liable for an additional amount, eighty percent of Montgomery’s uncollectible twenty percent. We agree and reverse the Court of Appeals holding that the city was liable for none of Montgomery’s twenty percent.
The statute required that, in rendering a verdict, the judge or jury determine what percentage of fault is ascribable to each party.26 At subsection 7, it stated:
[154]*154Notwithstanding subsection (3),[27] a governmental agency, other than a government or medical care facility, shall not be required to pay a percentage of any uncollectible amount that exceeds the governmental agency’s percentage of fault as determined under subsection (1). [MCL 600.6304(7); MSA 27A.6304(7).]
The expression “a percentage of any uncollectible amount that exceeds the [defendant’s] percentage of fault as determined under subsection (1)” appears also in subsection 6. There, the meaning of the language is clear: Where a party’s share of the obligation is uncollectible from that party, it may be reallocated among the other parties according to their respective percentages of fault. Each must then pay a percentage of the uncollectible amount equal to its own percentage of fault.
We conclude that the language in subsection 6, which is repeated in subsection 7, has the same meaning in both subsections. We find it incredible that the Legislature would use the same language in two successive sections to carry different meanings.
It is apparent that subsection 7 was included in the statute to limit the liability of governmental agencies, excepting hospitals and health care facilities. Without it, a governmental agency could be hable for as much as one hundred percent of any uncollectible amount. That is because subsections 5 and 6, which limit a party’s liability for an uncollectible amount to its per[155]*155centage of fault, do not apply where the plaintiff is not at fault.
In the present case, the jury found the plaintiff to be not at fault. Consequently, subsection 7 serves to apply the same liability and limitation of liability to the city as subsections 5 and 6 would have applied had plaintiff been at fault. This is evidenced by the Legislature’s use, in subsection 7, of the same language as appears in subsection 6.
Thus, we reverse the Court of Appeals decision regarding the apportionment of the verdict and hold that the city does bear a reallocation burden equal to its initial percentage of fault. The city is liable for eighty percent of the verdict plus eighty percent of the twenty percent ascribed to defendant Montgomery.
STATUTORY SETOFFS TO THE COLLATERAL SOURCE RULE
Plaintiff recovered from collateral sources, including worker’s compensation, uninsured motorist carrier, and social security. The collateral source setoff rule, MCL 600.6303(1); MSA 27A.6303(1), states in pertinent part:
In a personal injury action in which the plaintiff seeks to recover for the expense of medical care, rehabilitation services, loss of earnings, loss of earning capacity, or other economic loss, evidence to establish that the expense or loss was paid or is payable, in whole or in part, by a collateral source shall be admissible to the court in which the action was brought after a verdict for the plaintiff and before a judgment is entered on the verdict. Subject to subsection (5), if the court determines that all or part of the plaintiff’s expenses or loss has been paid or is payable by a collateral source, the court shall reduce that portion of the judgment which represents damages paid or payable by a [156]*156collateral source by an amount equal to the sum determined pursuant to subsection (2). This reduction shall not exceed the amount of the judgment for economic loss or that portion of the verdict which represents damages paid or payable by a collateral source.
Under subsection 6303(2), before judgment is entered, the trial court must determine what expenses have been paid or will be paid by a collateral source. The paid expenses are then applied to reduce the economic damages component of the verdict. Under subsection 6303(3),28 the plaintiff has ten days to notify all contractual lienholders of the verdict, and then the lienholders have an allocated time to exercise or lose their subrogation rights.
In Rogers, plaintiffs did not notify all contractual lienholders within ten days of the verdict. They claimed to have given notice before the verdict. Defendants sought setoffs, but the trial court refused them and granted plaintiffs leave to notify contractual lienholders within ten days of the judgment.
Nothing in MCL 600.6303(3); MSA 27A.6303(3) expressly forbids a trial judge from expending the period or recognizing “notice” occurring before the verdict. The judge did not abuse his discretion in this case. The statute also does not deprive a trial court of the authority to recognize that substantial compliance occurred in this case. The law does not require that a [157]*157lienholder lose its substantive rights. It does not provide a tortfeasor with a windfall, just because lien rights were exercised at a different time than within the statutory ten-day window.
A statute is to be construed to avoid inflicting hardship or reaching an unjust or unreasonable result. Franges v General Motors Corp, 404 Mich 590, 612; 274 NW2d 392 (1979). Here, we conclude that the overall statutory purpose was fulfilled: to avoid giving plaintiff either a double recovery or a double liability. To that end, the spirit and purpose of the legislation must prevail over the strict letter of the law State Treasurer v Wilson, 423 Mich 138, 144; 377 NW2d 703 (1985).
Therefore, we reverse the decision of the Court of Appeals and hold that the circuit court properly ruled concerning the collateral source rule.
CONCLUSION
We decline to overrule or modify our decision in Fiser. When legislative enactments are clear on their face, the role of the judiciary is not to articulate its view of policy, but to apply the statutes in accord with their plain language. Lorencz, supra. We conclude that any limitation in liability in police chase cases must come from the Legislature.
Thus, on the basis of Fiser, we affirm the decision of the Court of Appeals in Ewing and hold that summary disposition was improperly granted to defendants. In Rogers, we affirm the Court of Appeals decision regarding all aspects of the case, including the governmental immunity claim. However, we reverse the Court of Appeals decision regarding the apportionment issue and the collateral source statute. The [158]*158city is liable for eighty percent of the verdict plus eighty percent of the twenty percent ascribed to defendant Montgomery. Any outstanding liens will be satisfied upon the payment of the judgment to plaintiff.
Mallett, C.J., and Brickley and Cavanagh, JJ., concurred with Kelly, J.