Two of the most popular hybrid ADR processes are med-arb and arb-med. These methods combine the flexibility of mediation with the certainty of arbitration. Despite their advantages, both processes have been criticized. This article considers all sides of the question, and helps readers mitigate the processes’ alleged weaknesses.

America has a reputation as the most litigious country in the world [1].  And yet, only about six-tenths of one percent of cases result in trial. [2] The majority of the remainder are settled through some form of alternative dispute resolution (ADR). For example, almost 50% of Fortune 1000 companies choose ADR as an initial conflict-resolution mechanism, using litigation only where necessary [3].

In order to facilitate settlement, America’s courts have introduced mandatory ADR as a routine part of the process, imposing uniform policies and rules. [4] For those not embroiled in litigation, however, ADR is a matter of contract, and parties are free to agree to any dispute-resolution method that suits their needs. 

Widespread and longstanding use of hybrid ADR processes
indicates their usefulness.

Sometimes parties use hybrid processes, drawing on the strengths of two or more. Two popular hybrid processes are med-arb and arb-med.  In med-arb, the parties engage in mediation first.  If that fails, the matter is arbitrated.  The arbitration may be done by the same neutral or by a second one.  In arb-med, the matter is arbitrated and the award is sealed. It is then mediated. If mediation fails the award is unsealed and entered. [5] Sometimes the term “med-arb “ is used to refer to both processes, and sometimes to a situation in which a trial judge explicitly presses for settlement before deciding a case.  As discussed below, each variation has differing risks and incentives.

San Francisco lawyer and arbitrator Sam Kagel is credited with “inventing” med-arb to resolve a nurses’ strike in the 1970’s. Kagel’s work did popularize it [6], But med-arb , in all its forms, is an ancient process,  In ancient Greek and Egyptian traditions, med-arb was the norm rather than the exception. [7] Likewise in Islamic legal cultures, judges (qadi) have been expected to push for agreement before proceeding to trial for the past 1400 years. [8]. Both med-arb and arb-med share virtues of other forms of ADR – parties can control their own fate, relationships can be preserved, the resolution is swifter and less costly than litigation, and confidentiality is maintained.  In addition, med-arb and arb-med both have the virtue of finality, and (where the same person conducts both phases) of time and cost savings because a second neutral need not “get up to speed” with the dispute. [9] Because of these advantages, more than 51% of today’s Fortune 1000 use some form of these processes. This represents an 11% increase from 1997. [10]  In the U.S., the med-arb process has been used successfully to resolve complex multimillion dollar corporate conflicts. [11]

In California, a mediator retained in an agricultural labor dispute must submit a report with binding recommendations to the Agricultural Labor Relations Board within 21 days of a failed mediation concerning the collective bargaining agreement.  This is clearly a mandatory statutory med-arb provision, though the statute does not call it that. [12]

And California appellate authority explicitly supports the use of med-arb in Lindsay v. Lewandowski. [13] In Lindsay, the parties entered into a “stipulated settlement agreement” which provided that Lindsay would pay Lewandowski $190,000, and that disputes would be subject to “binding mediation”– a phrase that could mean med-arb. The weakness of the agreement at issue in Lindsay was that the parties failed to specify what the process entailed.  The court refused to imply one, holding instead that there was no meeting of the minds.    This ruling teaches that the decision to select med-arb, and the process to be used, must be clearly expressed. Nevertheless, the opinion supports the use of the ‘same-neutral med-arb process.  The Lindsay court expressly said:

“We do not suggest that parties are prohibited from agreeing that, if the mediation fails, they will proceed to arbitration. And, should they so desire, they may agree that the same person may first act as mediator and, if he or she fails in this task, act as the arbitrator. Whether or not this arbitrator (nee mediator) may consider facts presented to him or her during the mediation would also have to be specified in any such agreement.” (Emphasis added) [14]

Use of these hybrid processes is not restricted to America. One Japanese study found that 99% of cases submitted to arbitration actually undergo “med-arb.” To some degree, these surprising results are structural. Japanese arbitrators are expected to attempt mediation of the disputes before them, and routinely do so. [15]   However, Professor Tony Cole points out that as Japan interacts more with Western businesses, med-arb practices increasingly conform to Western norms. [16]

In other Pacific Rim Nations, such as China [17] Hong Kong, Singapore, [18] and Australia [19] disputants also frequently embrace med-arb. In Canada today, med-arb is used routinely to resolve public sector and other disputes.[20]  And  Professor Yasunobu Sato notes that the World Intellectual Property Organization –an agency of the United Nations addressing patent, trademark, copyright and similar issues – actively embraces med-arb [21]

One caveat here is that med-arb / arb-med in other cultures doesn’t always look the same as it does in the United States, though it bears the same name.  China, for example, has little stand-alone mediation little stand-alone mediation but much med-arb. [22] And the process is braided, rather than compartmentalized. That is, mediation occurs in stages throughout the arbitration. [23] Also, in China the Western concept of med-arbiter impartiality is alien; the ideal mediator is a respected authority figure with ties to both parties.[24] However, Professor Cheng [25] warns that it is dangerous to generalize about cultures, as modern attitudes change and cultures meld in the face of ongoing globalization.


Despite the popularity and obvious usefulness of med-arb, it remains controversial, with some arbitrators treating it as “heretical and even unethical.” [26] Even in the face of long previous experience, these critics believe the hybrid processes are inescapably unfair and coercive given the conflict between the neutral’s initial role as a facilitator and her potential role as an adjudicator. [27] These criticisms are discussed below:

Natural justice

Most criticism related to med-arb is aimed at the ‘same-neutral, mediation-first’ variety. The two main criticisms relate to due-process (sometimes referred to as natural justice) [28] and the potential for coercion. The most frequent criticism of the mediation-first variant of med-arb is that the neutral hears evidence in private caucus that the non-caucusing party cannot respond to, because he has not heard it.   In stand-alone mediation this is not a concern, because the neutral is not a decision-maker. In med-arb, evidence heard in mediation caucus is problematic if the matter goes to arbitration, because a party who was excluded from caucus during mediation has not had notice of “new evidence” presented during that private session, or an opportunity to respond to it. As Telford points out, concerns about exposure to improper evidence reflect doubts about the neutral’s ability to disregard that evidence, rather than a flaw in the med-arb process itself. [29]

Admittedly, although judges, arbitrators and jurors are expected to ignore inadmissible evidence, some research demonstrates that they have trouble doing so. This includes proposed settlement offers disclosed in mediation. [30] Other scholarship challenges these findings, indicating that curative instructions are effective, though imperfectly so. [31]   Megan Telford, who surveyed med-arbiters working in Ontario’s Grievance Settlement Board, reports that while the natural justice / due process issues raised earlier were a concern, the great majority of the neutrals surveyed  felt able to disregard inadmissible evidence.[32]   Similarly, Peter Robinson of Pepperdine’s Straus. Institute for Dispute Resolution reports that ninety-five  percent of California judges surveyed were unconcerned about hearing inadmissible information in mediation [33], apparently believing they could disregard it.    In the face of this uncertainty, some neutrals have addressed the natural justice issue by refusing to caucus during the mediation stage of a med-arb. [34] Indeed, in some jurisdictions caucusing in med-arb is not allowed. [35] In others, anything material to the arbitration that was revealed in caucus must be disclosed. [36]


The court in Lindsay v. Lewandowsky, supra opined that any med-arb agreement should specify whether or not the neutral may consider data learned in mediation when acting as an arbitrator.  As discussed above, there is some doubt whether disregarding information is even possible. The issue is important because communications in mediation may be statutorily privileged.  Generally, information from mediation is not to be used or disclosed in arbitration or trial. [37] All 50 states and the District of Columbia protect mediation privacy, either by statute or court rule. [38] Federal courts take a case by case approach under FRE sections 501 and 408. [39] Not all federal jurisdictions recognize the privilege [40], and the contours of state law and related court rules vary. So it’s imperative to check privilege issues in your jurisdiction before the med-arb session.   Under California state law, for example, all participants in the mediation including counsel and the mediator hold the mediation privilege, [41] So all must participate in a written or recorded [42] waiver if mediation information is to be introduced in the arbitration phase.    If an appropriate waiver is not executed, the arbitration award may be vacated. [43]

While a reasonable argument might be made that a med-arb participant was barred by implied waiver or estoppel from claiming wrongdoing, having agreed in advance to a procedure that inevitably put the arbitrator in possession of privileged information, the California courts have rejected the use of implied waiver or estoppel in the mediation confidentiality context, imposing an explicit waiver requirement in the face of inflexible statutory language [44].
Sample Med-Arb Waiver Agreement


Some critics of the same-neutral and judicial variants argue that neutrals coerce the parties into settlement by foreshadowing negative results in the adjudicative phase.  Such a scenario is plausible, if one ignores the reality that professional mediators do not achieve settlement “at all costs.” A neutral’s biggest asset is a reputation for fairness, without which they will get no business Both Telford’s Ontario mediators and Robinson’s California judges [45] report being carefully less evaluative in the mediation phase. In fact, parties are more motivated to settle in the shadow of arbitration.[46]  But this is a natural outcome of the process. There are likely to be hints of what is to come in any setting in which a third party decides. Going through the pretrial process often tells counsel how a trial will go – how the court will rule on certain pivotal evidence or argument.

The parties chose med-arb because they wanted the certainty of a result; otherwise, they would have chosen pure mediation. Even outside the med-arb process, a looming trial or upcoming arbitration encourages settlement. Therefore, what critics call “coercion” is actually a feature of any adjudicative process.

Likewise, critics complain that med-arb clients inflate the merits of a case in mediation in order to look good to the arbitrator. This could happen in med-arb, but is not limited to that context. “Spinning” one’s case happens before juries and neutrals in stand-alone processes as well.   In fact, there is no evidence that or posturing is more likely in the med-arb context than it is in single-process ADR, such as mediation or arbitration.


Some critics of med-arb suggest that participants are unlikely to share facts with the mediator that might reflect negatively on them, in view of his or her possible role as an arbitrator. In reality, participants are more forthcoming, more creative, and take more of a problem-solving approach in med-arb than in other contexts, perhaps out of a desire to avoid annoying the potential arbitrator.[47]  Even if there is non-disclosure, this situation is hardly unknown to experienced mediators.  It is not unusual for counsel to withhold extremely useful information – data that would probably settle the case -- so that he or she can “save it for trial.”  This occurs even if the point is expressly made that the settlement process is designed to avoid trial altogether.



The arb-med variant puts arbitration first. The award is sealed, and the mediation proceeds, with the award unsealed if there is no settlement.  The process is very effective.  Several studies report higher rates of settlement with arb-med than with other processes [48]. With the arbitration done, disputants can endeavor to settle with the details of that proceeding fresh in their minds, and no doubt with their predictions of the contents of the sealed envelope weighing heavily on their offers, and the likely risks sharply defined.

Disputants often settle late in the process – having gained the most complete and accurate information about value by waiting.  An arb-med participant has waited as long as possible and has the most information – trial is over. All that remains is to open the envelope. It is no wonder that arb-med is effective, as the parties struggle to avoid immediate exposure to perceived risks.

Like med-arb, arb-med is criticized as coercive.  And like med-arb, that coercion is an inevitable part of the process. With arbitration first there are no mediation-privilege or natural justice concerns, Arb-med does, however, front-load the expense. The cost of arbitration is wasted if a mediation-first approach would have resulted in settlement.  

There are endless varieties of med arb, some of which are summarized in table form in Med-Arb and it's Variants: Ethical Issues for Parties & Neutrals by Richard Fullerton. [49] It is up to the disputants and their counsel to craft a process appropriate for their dispute.


[1] E.g. Lipsett, S.M. American Exceptionalism: A Double-Edged Sword (1996), at p. 49

[2] Langbein, J. H.122 Yale Law Journal 522, 524 (2012); available at (last visited June 9, 2013)

­[3]  Stipanowich, T. J. and Lamare, J. R.,” Living with 'ADR': Evolving Perceptions and Use of Mediation, Arbitration and Conflict Management in Fortune 1,000 Corporations,” at Table A. Harvard Negotiation Law Review, Forthcoming; Pepperdine University Legal Studies Research Paper No. 2013/16. (February 19, 2013) Available at SSRN: or, (Last visited June 10, 2013).

[4] Participation in ADR is frequently mandatory in currently litigated cases.  See, e.g. Local Rule 16-15, United States District Court, Central District of California. (last visited June 10, 2013).

[5] Limbury, A. L. “Med-Arb, Arb-Med, Neg-Arb, and ODR (2011) (last visited June 10, 2013)

[6]Telford; Megan Elizabeth, “Med-Arb:  A Viable Dispute Resolution Alternative,” IRC Press, January,  2000.

[7] Roebuck, D. “The Myth of Modern Mediation.” Vol. 73 no. 1  Arbitration: The Journal of the Chartered Institute of Arbitrators (2007) pp. 105-116;

[8] Rashid, S. K., “Peculiarities & Religious Underlining of ADR in Islamic Law.” (2008)

[9] See Limbury, fn. 5, supra.

[10] Stipanowich, T. J. and Lamare, J. R., supra. n. 3 at p. 29

[11] Elliot, D. C.  34 Alta. L. R “Med/arb: Fraught with Danger or Ripe with Opportunity?” 163 (1995) (last visited June 15, 2013),

[12] CA Labor Code section 1164{d}

[13] 139 Cal. App 4th 1618

[14] Id. at 1625

[15] See Sato, Y., COMMERCIAL DISPUTE PROCESSING AND JAPAN, 283-286; See also Van Soye, S. C. “Med-Arb  Might be Right for Your Business Dispute.” at fn. 4 (last visited June 14, 2013),


[17] Cole, fn. 9,.at 52, 113.

[18] Stipanowich, T. J.  Yang, J. Jay Welsh, Chen Qiming, Peter Robinson, Tan Jinghui, Chen Guang, Jeff Kichaven, Denise

Madigan, Wang Hongsong, and Zhang Jianhua, “East Meets West: An International Dialogue on Mediation and Med-Arb in the United

States and China,” 9 Pepp. Disp. Resol. L.J. (2009) available at (last visited June 12, 2013)

[19] Sato, Y. “Hybrid Dispute Processing in Japan: Linking Arbitration with Conciliation.” Dispute Resolution Across Systems (Bhatia et al. , eds, 2008) 53, at 63-64.72  See WIPO Mediation Rules 13(b)(iv) (last visited June 12, 2013)

[20] “Med-Arb: Combined Mediation/Arbitration” (last visited June 14, 2013),

[21] See Telford, supra n. 6; See Elliot, supra n. 11

(22] See Sato, n. 291, supra

[23] Stipanowich, et al, n. 18, supra

[24] Law, Siew-Fang “The construct of neutrality and impartiality in Chinese mediation.” 22 Alt. Disp Resol, J.118, at 120

[25] See Cheng, Tai-Heng, “Reflections on Culture in Med-Arb’ at p. 426   (last visited June 15, 2013)

[26] Id.

[27] Phillips, G. F. “Same-Neutral Med-Arb: What Does the Future Hold?” 60 Disp. Resol. J. 24, at 26 (2005)

[28] Although some use the term “med-arb” to refer to two-step processes in which different neutrals preside, unless otherwise noted, the discussion will refer to same-neutral med-arb, which is the focus of most criticism.

[29]”Natural justice” is a concept used in the British Commonwealth. It is similar to what Americans think of as “due process”   For a brief discussion of basic principles, see “Natural Justice/ Procedural Fairness.  (last visited June 16, 2013)

[30] Telford, n. 6, supra.

[31] Telford, n. 6, supra. (judges); Tanford, J. A., “Thinking about elephants: admonitions, empirical research and legal policy,” 60 UMKC L Rev. 645 (1991) (jurors)

[32] Sklansky. D. A., “Evidentiary Instructions and the Jury as Other.” 65 Stan. L. Rev. 407, 425-437 (March 2013)

[33]Stipanowich, et al, n. 18, at p. 25

[34] As Telford points out, the clients of the Ontario Grievance Settlement Board are generally sophisticated “repeat players” (unions and employers) who could refuse to use a given arbitrator in the future if he or she was seen as biased.  Further, violation of the principles of natural justice or due process through the use of evidence obtained in mediation would be an independent ground to vacate the award Since Telford reports that the Grievance Settlement Board runs smoothly.  So such challenges must be few and far between.

[35] See Cal R. Ct. 3.823
 [36] See Limbury, n. 5, discussing the Commercial Arbitration Act for the Australian state of New South Wales. (COMMERCIAL ARBITRATION ACT 2010, SECTION 27D) 
[37] See, e.g., CA EVID. CODE 1119 (a)

[38] Deason, E.” The Quest for Uniformity in Mediation Confidentiality: Foolish Consistency or Crucial Predictability?” 85 Marq. L. Rev. 79 (2001) Professor Deason refers to “49 states and the District of Columbia,” excluding Delaware.  Delaware’s rules have since changed to conform to the majority. See, e.g. DEL. CH. CT. R. 174(c)(4)

[39] Levin, B. “Protecting the Goal of Mediation: Rule 408 & The Creation of a Mediation Privilege Under Rule 501” (last visited June 16, 2013)

[40] Imwinklereid, E.J. “The Path toward a Federal Mediation Privilege: Approaches toward Creating Consistency for a Mediation Privilege in Federal Courts (last visited June 16, 2013)

[41] See CA EVID. CODE 1122

[42] The recording must be reduced to writing within 72 hours. See CA EVID. CODE 1118

[43] See CA EVID. CODE 1128

[44] Simmons v. Ghaderi, 143 Cal. App. 4th 410 (Cal App. 2d Dist.) (2006)

[45] See nn. 6 and 33, supra.

[46]Telford, n. 6, supra.

[47] McGillicuddy, N. B.; Welton, G. L.; Pruitt, D. G. “Third-party intervention: A field experiment comparing three different models.”

53 Journal of Personality and Social Psychology, (Jul 1987) at pp. 104-112

[48] Id.; See Conlon, D. E., Moon, H., and K. Y. Ng,, “Putting the cart before the horse: The unexpected benefits of arbitrating before mediating. Paper presented at the61st Annual meeting of the Academy of Management, Washington, DC. (August 2001); Stipanowich, n. 18. Supra; Stipanowch, reports settlement in every arb-med in which he has presided.

[49] Table taken from Fullerton, R. “Med-Arb and Its Variants: Ethical Issues for Parties and Neutrals”, 65 DISP.RESOL. J. 52, at 59 (Oct 2010), available at (last visited 6/18/13)


The Agency for Dispute Resolution (ADR) is a premier provider of private dispute resolution services with a distinguished panel of neutrals. They specialize in administering effective mediation & arbitration of simple to complex civil disputes & commercial cases.