Top Menu

Examining the Minnesota Orchestra Redline Agreement Part 2

This installment of our ongoing examination of the Minnesota Orchestra Redline Agreement (MORA) will focus on proposed changes to Individual Contracts. All things considered, Individual Contracts are a mystery to just about everyone outside of those directly involved. Consequently, let’s change that by shedding some light on how all of this works then jump into examining the related MORA propositions.

Introduction Material

If you’re new to this series, please take a moment to expand this content which explains what a redline contract is, the ground rules for this examination, define related acronyms, and the overall goal of the series. You’ll also have an opportunity to download a complete copy of MORA.

However, if you’ve been following along since the onset, feel free to skip ahead.

EXPAND THE INTRODUCTION MATERIAL.

The Goal

What I hope to accomplish with this series is to provide a broader and more thorough understanding of the dynamic issues that go into a labor dispute. In the end, it’s much more than just compensation, benefits, musician compliment, and season length.

We’ll try to understand why clauses exist in the first place and what each side perceives as beneficial or injurious by modifying, eliminating, or leaving it untouched. Likewise, we’ll step outside of the world of sound bites, spin, and emotionally charged rhetoric by embracing a practical, evenhanded examination of the specific contractual language at core of this dispute.

[social-download button_id=”6fa40a3de4cf44280″ dl_id=”http://www.adaptistration.com/wp-content/uploads/MO-Redline-Agreement.pdf” theme=”blue” message=”Get social to download a complete copy of the MO redline agreement.” facebook=”true” likeurl=”CURRENT” google=”false” tweet=”true” tweettext=”I just downloaded the @mn_orchestra redline agreement at @adaptistration!” tweeturl=”CURRENT” follow=”true” linkedin=”true” linkedinurl=”CURRENT”/]

Ground Rules

  1. Avoid universal application. Perhaps one of the most important bear traps to avoid throughout this process is to assume that these issues are universally applicable to all orchestras; simply put, they aren’t. Just because the Minnesota Orchestra does something, doesn’t mean the same should hold true somewhere such Omaha and vice versa.
  2. Understand the acronyms.
    1. MOA = Minnesota Orchestra Association (the board and/or executive management)
    2. MMO = Musicians of the Minnesota Orchestra
    3. MORA = Minnesota Orchestra Redline Agreement (the version submitted to musicians as their last official offer and subsequently voted down on 9/29/12)
    4. CBA = Collective Bargaining Agreement (also known as “the contract,” “master agreement,” or simply “agreement”).
  3. First hand clarification. When possible, additional insight, justification, and rationale behind why changes have been presented and/or why changes are opposed will be provided by official MOA and MMO spokespersons. If such information is provided after an article is published, it will be included as an article update.
  4. Irreproachable comments. Comments are always welcome but readers are encouraged resist the temptation to submit a comment when upset. If you aren’t already familiar with Adaptistration’s comment policy, please take a moment to review (via the “Blog Policy” tab).
  5. Interconnectivity. Although we will be examining groups of related terms in each installment, it is important to remember that very few of these issues exist in an institutional vacuum; meaning, just because correlations aren’t made between one or more contractual items doesn’t mean they don’t exist. It is recommended that following each installment, readers think back to previous posts and attempt to identify any potential connections.

What Is a Redline Contract?

Simply put, a redline contract (sometimes called markup or strikeout) is a version of proposed modifications that show:

  • The original language.
  • Modifications in the form of strikeout and bold formatting; the former being a removal with the later an addition or modification.

Depending on where the revision process is, redline contracts come in varying degrees of detail but it is normal for final versions to include every possible change, right down to modified section and subsection numbering. The MOA redline agreement used for these purposes is an example of the latter.

Ignoring The Inconsequential Stuff

In order to establish clear parameters, let me be clear that we will be skipping over housekeeping markups, such as changes to article numbers, dates, and all of the bits and pieces that need to change.

MORA 1 - dates

We’re not going to sweat the little things.

Professional Disclaimer

search

Over the past two decades, a sizable portion of my consulting work has revolved around individual contracts. My position is a bit unique in that I work on both sides of the fence; meaning, I represent both musicians and managements (although the latter is related to providing counsel and conducting research).

Since all of my work in these matters is conducted under strict non disclosure agreements, I will not be providing direct examples. Nonetheless, my work has involved securing terms for more than 100 concertmasters, principals, assistant principals, associate principals, and fixed chair section musicians in the largest budget orchestras down through those with annual budgets of a few million dollars.

An interesting side note here: individual contracts are quite similar to the individual employment agreements crafted by orchestra executives that define not only base compensation but the array of benefits and perks above and beyond the minimum package offered to all administrative employees.

The preparation, method, and process (sometimes better defined as posturing) involved is nearly identical to the way individual musician contracts are crafted. Having worked for executives seeking to secure terms, boards seeking to get executives to work for less, musicians seeking improved terms, and executives tasked with minimizing overscale costs, I can firmly attest to an abundance of ironic similarities in the process regardless which parties are involved.

Individual Contracts Defined

Individual Contracts, also referred to as overscale contracts, are employment terms negotiated between an individual musician and the employer. Here are some key points to keep in mind:

  • They do not supplant terms specified in the collective bargaining agreement.
  • They may not contradict terms in the CBA and musicians with individual contracts are afforded all of the same protections and are responsible for all of the same duties and responsibilities as defined in the CBA.
  • An employer is under no obligation to offer or provide for individual contracts except in instances where minimum overscale terms for specific orchestral positions are defined in the CBA (more on that below).
  • Monetary terms are typically defined by dollar value and not by percentages over and above the base scale, except in instances where minimum overscale terms are defined in the CBA.
  • There is no universal formula for determining individual contract rates and the amount orchestras allocate to this expense varies wildly amid groups in the same budget class as well as those in different classes (oh, the stories I could tell).
  • Musicians may negotiate for monetary and non-monetary terms.
  • Individual contracts may have fixed or “evergreen” terms.
  • Some organizations allow individual contracts to be negotiated or renegotiated at any point where others create specific windows of opportunity.
  • Some organizations require musicians to negotiate directly (meaning, they prohibit the use of outside negotiators) while others have no restrictions.
  • When bargaining for individual contract terms, anyone from the executive administrator down through a mid level operations manager may represent the association; however, final terms are typically approved by the executive.
  • Music directors and board members are not actively involved in the bargaining but his/her opinion regarding the musician involved can have a great deal of influence in the outcome. Likewise, it is typical for management to confer with the music director in order to determine the value of any given musician.

Simply put, there’s no prescribed method for arriving at optimum individual contract conditions; instead, terms are determined more by individual tenacity, competition, accomplishment, and the respective organization’s labor culture.

In order to acquire a clear understanding, it’s critical to point out that some orchestras set fixed parameters for monetary terms (especially middle and lower budget groups); meaning, an individual musician can negotiate up to, but not past certain caps. Similarly, those parameters can provide a minimum starting point for overscale but not place any caps. In short, the sword can swing both directions.

However, it’s worth pointing out that mandated overscale defined in the collective bargaining agreement is typically restricted to monetary compensation and individual musicians are still free to negotiate non-monetary terms, some of which incur no costs to the organization.

Article III; Section 3.1 and 3.2

MORA 1 Individual Contracts Section 3.1 and 3.2

MORA 1 Individual Contracts Section 3.1 and 3.2

Quantitative Summary

Language acknowledging the existence of individual contracts is quite common. In some cases, it is a simple sentence or two indicating requests for individual contract terms are or are not entertained and if so, the related conditions. In most cases, it’s common to provide some parameters such as periods when bargaining may or may not occur, such as a prohibition during active CBA bargaining.

The most profound change in the MORA proposal is the addition of language that terminates all individual contracts by September 17, 2017 or earlier for any agreements with fixed expiration dates occurring before that threshold.

Although the MORA proposal doesn’t prohibit individual contracts, it is clear that the orchestra intends to trim back on the quantity and/or quality of terms. This is evidenced by the removal of a provision in Section 3.2 that requires the MOA to honor the terms of an existing individual agreement should both parties fail to reach mutually agreeable terms during any subsequent renegotiation.

Other proposed changes include removing deadlines for both parties to provide terms and official replies. The only language to remain unchanged is related to the termination individual contracts.

Management Perspective

On face value, the rationale for including language that guarantees the organization will shed itself of individual contract terms while simultaneously proposing steep cuts to base pay seems self evident. Since most individual contracts define dollar values as opposed to percentages over base compensation, the need to remove any language that protects those terms would be critical unless the cost differences have been accounted for by increasing the earning gap between musicians on the top and those earning entry level wages.

Eliminating guarantees for renegotiating existing individual contracts only increases the likelihood that the only time a musician will ever be able to bargain for individual terms is at the point s/he is hired and/or receives tenure. This would produce much lower long term costs as well as prevent any musician from attempting to leverage an offer from a peer orchestra for the purpose of securing better terms to remain with the ensemble.

Musician Perspective

For employees assuming duties and responsibilities that exceed those of standard members, individual contracts are the only option for generating additional benefits that acknowledge those contributions.

Consequently, the musician perspective is comparatively cut and dry and it provides freedom for those key musicians to obtain a degree of acknowledgement and recognition for their role.

Again, it’s fair to point out that generalities on these matters shift from one budget class to the next but for a group the size of the Minnesota Orchestra, it is almost unheard of for the musicians to adopt measures via the CBA that would limit or otherwise cap individual contract terms.

Observations

Although the proposed language in this section of MORA may seem like a straightforward effort to reduce overall costs, it also contains the potential to produce a considerable amount of dynamic consequences; not the least of which includes:

  • Increased workload. Unless the MOA is planning to deny most individual contract requests out of hand, they can expect an increase in the number of musicians initiating bargaining demands. Likewise, removing some of the existing protections (such as continuing previous conditions if new terms aren’t mutually agreeable) can actually work against management in that it decreases employee complacency. As a result, the organization can expect an increase in the number of bargaining requests along with a higher degree of antagonism.
  • Increased turnover. Regardless of how much talk exists about an oversupply of musicians, that hasn’t made a noticeable dent in individual contract terms and overscale. Orchestras continue to struggle to attract and retain key artistic employees. By removing the ability to renegotiate terms mid-contract, the MOA would position itself as a very “poachable” ensemble.
  • Decreased quality. Although musicians tend to be fond of defining an organization’s overall attractiveness by the level of base salary, that’s only accurate in an applicable sense for smaller and mid budget size orchestras. The larger the budget, the more individual contract terms impact an orchestra’s artistic gravity. In short, do you really think it matters to the potential principal oboe if Orchestra A’s base salary is three percent less than Orchestra B? Not a chance; instead, it is individual contract terms that end up tipping the scales.
  • Guaranteed cyclical bargaining. By proposing the term “All individual contracts will terminate on or before the expiration date of this agreement” without an indication that the term applies solely to the 2012-2017 agreement, existing and incoming musicians eligible for individual contract improvements can expect those terms to be terminated at the end of every CBA. Thus initiating the individual contract bargaining cycle from scratch along with each new CBA, not to mention adding another layer to the “increased workload” point above.

In Closing

It is worth noting that when negotiating monetary terms for individual contracts; managers typically refuse to link overscale payments to base compensation as well as resist terms that include minimum percent increases from one season to the next.

Traditionally, this provides a degree of insulation over ancillary increases in overscale expenses since base compensation rises at more or less steady rates; while individual musicians are less likely to engage in bargaining for new terms once their initial individual contract is settled.

In an ironic twist, this approach works against the MOA and it serves as reasonable grounds for why MORA contains what might be considered counterproductive Individual Contract terms. If the organization had linked individual contract monetary terms as a percent over and above base salary, they would have been guaranteed parallel cost controls for whatever terms the musicians ultimately accept.

Assuming the MOA’s goal isn’t to deliberately hasten the turnover of key musicians, unilaterally dissolving all overscale agreements via CBA decree is a high risk approach to dealing with what the organization defines as long term financial problems.

In the end, this peek inside the much larger proposal will prove to be a useful when developing your own practical on the dispute. In the meantime,  if you have any questions about individual contracts in general or the MORA proposal, send them in via a comment below.

Looking for more MORA discussion? Check out the latest article on the topic from conductor Bill Eddins over at Sticks and Drones.

Trackbacks/Pingbacks

  1. Adaptistration Turns Nine! | Adaptistration - November 8, 2012

    […] after the last Minnesota Orchestra Redline Agreement (MORA) installment, both of my colleagues, who I’ve known for years, wrote in to say they had zero clue that […]

  2. Writer’s blockheads. | non divisi - July 5, 2013

    […] the master agreement, starting from scratch in terms of industry-standard working conditions (check this out for starters). Then they walked away and literally locked the doors. In essence, the MOA […]

Leave a Reply

Navigation