By Robert D. Owen, a Partner with Eversheds Sutherland (US) LLP in the firm’s New York, NY office and President of the Electronic Discovery Institute, and Francis X. Nolan, an Associate with the firm.

Economics teaches that underpriced commodities are inevitably overconsumed.  This truth has become increasingly evident in the world of large-litigation discovery over the last decade.  Requesting parties (typically plaintiffs) are not effectively limited by outside forces to moderate the breadth of the discovery, nor are they incentivized to curb excess demands.  Discovery is currently free to requesting parties, and judges have limited time to mediate and resolve inevitable discovery disputes.  Accordingly, discovery has spun out of control.

The Advisory Committee on the Rules of Civil Procedure (“Committee”) acknowledged the problem of runaway discovery in the 2015 amendments, when it expressly incorporated the proportionality factors into Rule 26(b)(1).  The Committee recognized that the world is simply full of more “relevant” information now than it used to be, and that courts and litigants require limits on what is discoverable.

Although the 2015 amendments have been largely successful, further amendments are necessary to reinforce “the common-sense concept of proportionality,” articulated by Chief Justice John Roberts in his 2015 Year-End Report on the Federal Judiciary.  One effective and simplistic way to achieve this goal is to incorporate into the discovery rules a percentage-based “requestor pay” provision.  This Legal Backgrounder first explains how litigation has reached the current tipping point, and then offers examples of recent attempts by the Committee and the judiciary to push back against rising discovery costs.  The publication concludes by offering a basic outline for a percentage-based requestor-pay rule, while also noting the advantages and hurdles attendant with such a proposal. 

The Evolution of Discovery

For centuries, civil litigation was resolved without parties gaining pretrial access to potentially relevant information.  In the absence of a formal discovery process, parties encountered their adversaries’ evidence at the trial.  Courts did not require a perfect record—always impossible—and disputes were resolved under the preponderance of the evidence standard.  Even now, records are never perfect and litigation is rarely won because the prevailing party was able to access every e-communication ever sent, or every draft of every financial report ever prepared.  For one thing, civil trials are increasingly rare, and even when cases go to trial, only a small fraction of the produced data is entered into evidence.  This is particularly true in larger cases.

The trend in the electronic era has been for parties in litigation to seek everything: all remotely relevant information, no matter how useful, simply because it is available.  And although the adversary system does not require perfection, lawyers by training and inclination tend to seek it, and judges understand the inclination and themselves want results to be as close to perfect as possible.  But perfection comes at a price.

In 1938, the first Federal Rules of Civil Procedure enshrined the then-radical philosophy of full pretrial discovery.  But at the time files for most cases could fit into a few folders or a single banker’s box and there was no such thing as “costs of discovery.”

For the ensuing four decades (and in some cases beyond), it was common practice for the requestor to pay production costs for discovery it sought.  Papers would be collected, piled on a conference room table for the requesting party to see, and the latter would pay for the photocopies of the clipped pages.  In 1978, the U.S. Supreme Court issued its opinion in Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 358 (1978), where it stated that “[u]nder [discovery] rules, the presumption is that the responding party must bear the expense of complying with discovery requests,” but the costs of production during that time were negligible.  Now they are not.

First and foremost, the rise (and now near-omnipresence) of electronically stored information has made printing costs of a few file folders a thing of the past.  Communications are now mostly electronic, and countless drafts of spreadsheets, charts, graphs, and tables have replaced the few hard copies in years past.  The sheer volume of searchable, discoverable, and admissible information is far greater than it used to be.  The costs of managing the duties of affirmative preservation, collection, review, and production are correspondingly higher.  This explosion in volume of relevant data has led to rising volumes and costs and an increasingly contentious discovery process.

Recent Attempts to Mitigate Burdensome Discovery

The rising costs of discovery are well-known to the Committee.  In their 2015 Notes to the Federal Rules of Civil Procedure, the Committee stressed, “Courts and parties should be willing to consider the opportunities for reducing the burden or expense of discovery as reliable means of searching electronically stored information become available.”  Note 16.  The Committee further allowed, through amended Rule 26(c)(1)(B), for “an express recognition of protective orders that allocate expenses for disclosure or discovery.”  Note 21.  But the Committee pulled its punches on the notion of cost-shifting or fee allocation, stating that although it recognized the authority of courts to allocate expenses, that “does not imply that cost-shifting should become a common practice.”  Ibid.  Most unfortunately, the Committee concluded by referencing the now nearly 40-year old Oppenheimer decision and stating, “Courts and parties should continue to assume that a responding party ordinarily bears the costs of responding.”  Ibid.

As the Committee noted, a minority of magistrate and district court judges have incorporated cost-shifting into their case management orders.  For example, Judge Paul Grimm of the District of Maryland adopted individual rules that divided discovery into two “phases.”  Phase 1 covers discovery “likely to be admissible” and thus “narrower than the general scope of discovery stated in Rule 26(b)(1).”  Discovery Order, Jan. 29, 2013 at 1 (Emphasis in original).  Discovery that falls within Phase 2 is more expansive but subject to different presumptions.  Requesting parties must make a showing of good cause that the discovery sought in Phase 2 is proportional to the needs of the case.  Furthermore, Judge Grimm requires that requesting parties show good cause as to why, if Phase 2 discovery is permitted, the requesting parties should not pay some, or even all, of the costs associated.

Judge Grimm, and others like him, should be commended for taking an active role in managing discovery, as encouraged by the 2015 Advisory Committee Notes.  But they remain outliers, and their case management orders can be difficult and burdensome to apply in practice.  For example, a showing of “good cause” to determine cost shifting, as required by Judge Grimm, implicates a subjective standard and can naturally lead to dreaded litigation on litigation.  On the other hand, an objective standard or formula for cost-sharing could be implemented at the early stages of discovery, thereby incentivizing parties to tailor their discovery demands to each case, and avoid a laborious two-step discovery process and time-consuming judicial intervention.

Other courts have implemented cost-sharing as a means of resolving discovery disputes.  In Lubber, Inc. v. Optari LLC, No. 3:11-0042 (M.D. Tenn. Mar. 15, 2012), an obviously exasperated Magistrate Judge Joe Brown directed parties in a nasty discovery dispute to share costs associated with producing discovery.  In that decision, Magistrate Brown opined on many of the same issues discussed here.

It is the Magistrate Judge’s experience and the view of a number of economists who have studied this issue that where the requesting party bears a part of the cost of producing what they request, the amount of material requested drops significantly.  When a party has to contemplate whether the last possible bit of information will cost them more than it is worth, they quit asking for items of marginal relevance.

Like Judge Grimm, Magistrate Brown’s approach to discovery disputes is in keeping with an equitable and “common-sense approach” to discovery.  But the decision in Lubber is also an example of what can happen when parties do not have prospective guidance—or guidelines—at the start of discovery.  In Lubber, Magistrate Brown was forced to step in as a referee to resolve a dispute that likely would not have escalated had the Magistrate been free to act as a discovery manager.  If the requesting parties had been required to pay even a small share of the costs associated with their discovery demands, their demands would have likely been more narrowly tailored, and the dispute could have been mitigated or even avoided altogether.  Again, Magistrate Brown wrote:

As long as requesting the last bit of information costs nothing they have little, if any, incentive not to request it.  Even if they choose never to look at it, they have put the opposing party to the cost of production.  In some cases discovery becomes a tool with which to bludgeon the other side into submission.

Under the current regime, there are few limits on what a requestor can get, free of charge, in discovery.  Those limits require that the responding party push back on the requestor, including through expensive litigation-on-litigation motion practice, and that a judge with limited time hear the motion and agree with the responding party.  Even in situations where a judge is actively involved in discovery from the outset of litigation, it is unrealistic to expect the judge to have a full command of all of the facts and nuances of a case to accurately rule on discovery issues.  Furthermore, discovery orders are almost never appealed, putting even more discretion in the hands of particular judges and magistrates.

There is another path forward.  

The Proposed Rule

What is the purpose of a “co-pay” requirement when patients go to their doctors’ offices?  Its purpose is to apply a small brake on a patient’s ability to have visit after visit.  It modestly deters patients from requesting services that would otherwise be free.

The proposal stems from the same rationale: requesting parties pay 10% of costs of production.  This reflects a straightforward, “common-sense” approach to resolving many of the issues that arise in discovery disputes.

It is unrealistic to expect that a full-blown requestor-pay rule will garner enough support to be implemented into the Federal Rules of Civil Procedure any time soon.  In fact, it is inadvisable because the rule could create the same perverse incentives in reverse: a producing party could engage in over-production as a tactic to starve its adversary of litigation funds.

Simply put, requesting parties should have some “skin in the game” when seeking discovery from their adversaries.  10% is a good start.  As any litigator will attest, writing even a small check can have a sobering effect on a party’s discovery.

Obvious questions arise.  Perhaps most prominent is: “how are costs calculated?”  Soft costs can include attorney and paralegal time other than review costs, vendor vetting (including for cybersecurity purposes), and coping with the risks of allowing data outside the firewall.  Hard costs can include fees for mining hard drives, accessing archives, restoring backup tapes, review costs, and hosting data for review and production.

Further, determining the costs themselves will lead to litigation on litigation.  A flat-fee approach to assessing costs can counter these potential problems.  One version of this—although others surely exist—would be a cost per-produced page.

Parties also will not be able to determine, at the outset of discovery, exactly how much data will be produced in response to their demands.  The parties could effectively deal with this through the meet-and-confer process, which would include the magistrate or district judge as a case manager.  During the meet-and-confer, the parties could estimate—based on the number of custodians, for example—how much data will be reviewed and ultimately produced, and then reach an agreement as to what limits would be put in place to prevent over-production.

Incorporating the 10% rule into the meet-and-confer process would give both sides an incentive to minimize discovery.  As it currently stands, one side has no such incentive and is content to impose costs on the other.  A requesting party would be motivated to refine and narrow the scope of data requested, and make their demands proportional to the case.  The process would help fulfill what Chief Justice Roberts, in his 2015 Report, called “the obligation of judges and lawyers to work cooperatively in controlling the expense and time demands of litigation.”

The proposed rule also offers a secondary backstop against overproduction, because the producing party would still pay 90% of the discovery costs, thereby incentivizing it to keep costs down and to not over-produce.  Further, a requirement of this nature would promote settlement of discovery disputes between the parties without judicial interference, thereby saving time, money, and resources, all without imposing disproportionate expense on the requestor.

This approach would be just as effective in small, relatively inexpensive litigation.  That said, a proposed requestor-pay rule could include an exemption for parties proceeding in forma pauperis.

With any change comes resistance, and this proposal will undoubtedly draw the ire of many a plaintiffs’ attorney in particular.  Implicit in the current structure, of course, are two assumptions: (1) plaintiffs are typically the parties making costly requests and (2) defendants are typically the deeper pockets.  The proposal would also benefit plaintiffs, however, because cases would be resolved more quickly and efficiently, and with fewer costly discovery disputes—which after all impose real costs on both parties.  All at the cost of only 10% of the production costs for the discovery they need to prosecute their claims.


Skyrocketing discovery costs call for a fundamental shift in how parties approach and pay for discovery.  The proposed 10% co-pay rule would incentivize more reasonable, tailored requests and encourage greater cooperation among the parties with less judicial intervention of disputes.  This would all occur without imposing disproportionate expense on requestors.  The most important results would include a reduction in the number of discovery disputes requiring judicial intervention, a concomitant reduction in the overall scope and cost of production, and a refocusing of attention on the primary legal and factual issues in each dispute.