Category: Copyright Office

Deduplication and Discovery: The Smoking Gun in the Machine

WINSTON

“Wipe up all those little pieces of brains and skull”

From Pulp Fiction, screenplay by Quentin Tarantino and Roger Avary

Deduplication—the process of removing identical or near-identical content from AI training data—is a critical yet often overlooked indicator that AI platforms actively monitor and curate their training sets. This is the kind of process that one would expect given the kind of “scrape, ready, aim” business practices that seems precisely the approach of AI platforms that have ready access to large amounts of fairly high quality data from users of other products placed into commerce by business affiliates or confederates of the AI platforms.

For example, Google Gemini could have access to gmail, YouTube, at least “publicly available” Google Docs, Google Translate, or Google for Education, and then of course one of the great scams of all time, Google Books. Microsoft uses Bing searches, MSN browsing, the consumer Copilot experience, and ad interactions. Amazon uses Alexa prompts, Facebook uses “public” posts and so on.

This kind of hoovering up of indiscriminate amounts of “data” in the form of your baby pictures posted on Facebook and your user generated content on YouTube is bound to produce duplicates. After all, how may users have posted their favorite Billie Eilish or Taylor Swift music video. AI doesn’t need 10000 versions of “Shake it Off” they probably just need the official video. Enter deduplication–which by definition means the platform knows what it has scraped and also knows what it wants to get rid of.

“Get rid of” is a relative concept. In many systems—particularly in storage environments like backup servers or object stores—deduplication means keeping only one physical copy of a file. Any other instances of that data don’t get stored again; instead, they’re represented by pointers to the original copy. This approach, known as inline deduplication, happens in real time and minimizes storage waste without actually deleting anything of functional value. It requires knowing what you have, knowing you have more than one version of the same thing, and being able to tell the system where to look to find the “original” copy without disturbing the process and burning compute inefficiently.

In other cases, such as post-process deduplication, the system stores data initially, then later scans for and eliminates redundancies. Again, the AI platform knows there are two or more versions of the same thing, say the book Being and Nothingness, knows where to find the copies and has been trained to keep only one version. Even here, the duplicates may not be permanently erased—they might be archived, versioned, or logged for auditing, compliance, or reconstruction purposes.

In AI training contexts, deduplication usually means removing redundant examples from the training set to avoid copyright risk. The duplicate content may be discarded from the training pipeline but often isn’t destroyed. Instead, AI companies may retain it in a separate filtered corpus or keep hashed fingerprints to ensure future models don’t retrain on the same material unknowingly.

So they know what they have, and likely know where it came from. They just don’t want to tell any plaintiffs.

Ultimately, deduplication is less about destruction and more about optimization. It’s a way to reduce noise, save resources, and improve performance—while still allowing systems to track, reference, or even rehydrate the original data if needed.

Its existence directly undermines claims that companies are unaware of which copyrighted works were ingested. Indeed, it only makes sense that one of the hidden consequences of the indiscriminate scraping that underpins large-scale AI training is the proliferation of duplicated data. Web crawlers ingest everything they can access—news articles republished across syndicates, forum posts echoed in aggregation sites, Wikipedia mirrors, boilerplate license terms, spammy SEO farms repeating the same language over and over. Without any filtering, this avalanche of redundant content floods the training pipeline.

This is where deduplication becomes not just useful, but essential. It’s the cleanup crew after a massive data land grab. The more messy and indiscriminate the scraping, the more aggressively the model must filter for quality, relevance, and uniqueness to avoid training inefficiencies or—worse—model behaviors that are skewed by repetition. If a model sees the same phrase or opinion thousands of times, it might assume it’s authoritative or universally accepted, even if it’s just a meme bouncing around low-quality content farms.

Deduplication is sort of the Winston Wolf of AI. And if the cleaner shows up, somebody had to order the cleanup. It is a direct response to the excesses of indiscriminate scraping. It’s both a technical fix and a quiet admission that the underlying data collection strategy is, by design, uncontrolled. But while the scraping may be uncontrolled to get copies of as much of your data has they can lay hands on, even by cleverly changing their terms of use boilerplate so they can do all this under the effluvia of legality, they send in the cleaner to take care of the crime scene.

So to summarize: To deduplicate, platforms must identify content-level matches (e.g., multiple copies of Being and Nothingness by Jean-Paul Sartre). This process requires tools that compare, fingerprint, or embed full documents—meaning the content is readable and classifiable–and, oh, yes, discoverable.

Platforms may choose the ‘cleanest’ copy to keep, showing knowledge and active decision-making about which version of a copyrighted work is retained. And–big finish–removing duplicates only makes sense if operators know which datasets they scraped and what those datasets contain.

Drilling down on a platform’s deduplication tools and practices may prove up knowledge and intent to a precise degree—contradicting arguments of plausible deniability in litigation. Johnny ate the cookies isn’t going to fly. There’s a market clearing level of record keeping necessary for deduping to work at all, so it’s likely that there are internal deduplication logs or tooling pipelines that are discoverable.

When AI platforms object to discovery about deduplication, plaintiffs can often overcome those objections by narrowing their focus. Rather than requesting broad details about how a model deduplicates its entire training set, plaintiffs should ask a simple, specific question: Were any of these known works—identified by title or author—deduplicated or excluded from training?

This approach avoids objections about overbreadth or burden. It reframes discovery as a factual inquiry, not a technical deep dive. If the platform claims the data was not retained, plaintiffs can ask for existing artifacts—like hash filters, logs, or manifests—or seek a sworn statement explaining the loss and when it occurred. That, in turn, opens the door to potential spoliation arguments.

If trade secrets are cited, plaintiffs can propose a protective order, limiting access to outside counsel or experts like we’ve done 100,000 times before in other cases. And if the defendant claims “duplicate” is too vague, plaintiffs can define it functionally—as content that’s identical or substantially similar, by hash, tokens, or vectors.

Most importantly, deduplication is relevant. If a platform identified a plaintiff’s work and trained on it anyway, that speaks to volitional use, copying, and lack of care—key issues in copyright and fair use analysis. And if they lied about it, particularly to the court—Helloooooo Harper & Row. Discovery requests that are focused, tailored, and anchored in specific works stand a far better chance of surviving objections and yielding meaningful evidence which hopefully will be useful and lead to other positive results.

Who’s Coming to Lunch? What Do Personnel Changes at Copyright Office Mean for MLC?

If you’ve been following the news lately, you’ll have heard that President Trump has made some personnel changes at the Library of Congress and the head of the U.S. Copyright Office (styled as the “Register of Copyrights”). When the dust settles we’ll see if these changes stick, but my bet is they probably will. This is because the President was probably within his authority to replace the Librarian of Congress (a presidential appointee). Remember that the Librarian is a “principal officer of the United States” who ultimately reports to the President. We’ll come back to that point.

Because the Librarian appoints the head of the Copyright Office for an unspecified term and can terminate that person, there’s probably an argument for the President being able to terminate the “Register” directly if there’s a vacancy in the Librarian’s office especially if there’s urgent business before the Copyright Office. Alternatively, there’s definitely an argument for the replacement Librarian, “Acting” or otherwise, to be able to terminate the non-Senate confirmed Register. (See a similar argument from Professor Volokh.)

So whatever the sequence, the result is likely the same. Was it prudent? No. Was it well-handled? No. Is it enforceable? Quite probably. That doesn’t mean that those who are terminated can’t or shouldn’t pursue claims, but I think it does mean that their respective replacements are going to take over. The topic that is front and center in most discussions of these movements is Big Tech’s lobbying on AI and that is well to be concerned about because today is Wednesday and Big Tech is still trying to screw us. In that regard it is a day like any other.

But there is other pending business before the Copyright Office that will now be supervised by a Department of Justice lawyer with an entirely different background and set of relationships to all prior Registers. My bet is that the culture at the Copyright Office is about to change. I would say change radically, but I’d be skeptical that anything in Washington changes radically. For example, remember that the Library of Congress/Copyright Office public database apparently uses an older Oracle database system and/or COBOL or PL/SQL for data processing.  The user interface is HTML with embedded JavaScript, and uses CGI or early Java-based web tools for query submission. That’s right–1998 technology. Helloooo DoGE.

One item of pending business is the 5-year redesignation oversight review of the MLC’s operations and a review of the MLC’s investment policy on the $1.2 billion black box (or more) that is gradually inching its way toward a market share distribution with little or no explanation.

For reasons known only to the lobbyists who wrote Title I of the Music Modernization Act, the Copyright Office was given oversight of the MLC and its hedge fund.   As anyone could have predicted who’d studied the culture of the Copyright Office for five minutes, that oversight is effectively meaningless.  The MLC has just refused to allow any transparency over their hedge fund—over a billion dollars of other people’s money—and the Copyright Office so far has let that happen.  As Guy Forsyth wrote, Americans are freedom loving people and nothing says freedom like getting away with it.

So there’s a deeper structural issue with the MLC’s oversight: the Copyright Office is required to review the MLC every five years, but it has no real enforcement powers other than refusing to redesignate the quango which would create a huge disconnect between the sunny narrative of aspirations for the “historic” Title I of the MMA that created the MLC and the dark underbelly of the utter failure of that legislation that no one talks about at parties. Unlike executive agencies like the DOJ, FTC or SEC, the Copyright Office can’t subpoena records, issue fines, or force compliance. Its first five-year review—launched in January 2024—is now grinding on in its second year, with no public recommendations or reforms issued to date despite the requirements of the moment.

With an emphasis on regulatory accountability, the Trump administration might push for more rigorous oversight of the MLC’s operations, including its data practices and how it invests the black box OPM funds. Oversight could be enhanced through a combination of Copyright Office audits and a potential executive branch role—such as a streamlined agency focused on government efficiency. The goal: protect creators’ money and ensure the MLC’s compliance without increasing taxpayer burden. Costs for such oversight could, and arguably should, be charged back to the MLC which is funded by the richest corporations in commercial history.

In fact, beefing up the Copyright Office’s oversight role may actually be required. As Professor Volokh observes:

The answer appears to be that the Library of Congress is actually an Executive Branch department for legal purposes [and not in the Legislative Branch], though it also provides some services to Congress. Indeed, I think it has to be such a department in order to have the authority that it has over the implementation of copyright law (via the Register of Copyrights): As Buckley v. Valeo (1976) made clear, in a less famous part of its holding, Congress can’t appoint heads of agencies that exercise executive powers.

Of course the Librarian has to be confirmed by the Senate, although under vacancies rules, an acting Librarian has pretty much the full authority of the office for 210 days without Senate confirmation. The Register is not Senate confirmed, so there’s an odd juxtaposition where Trump’s Acting Librarian could be replaced, but the Register is not subject to the 210 day clock.

This is all down in the weeds in Appointments Clause land. But you get the idea. Paul Perkins, who was serving as an Associate Deputy Attorney General at the U.S. Department of Justice, will soon be looking at the MLC. My understanding is that Mr. Perkins is the deputy of Todd Blanche, who is now taking over as acting Librarian. (Todd Blanche who currently serves as the 40th United States Deputy Attorney General, having been confirmed by the Senate. He was formerly a partner at Cadwalader and former federal prosecutor in the SDNY.)

And just wait til DoGE gets a load of that COBOL programming and a billion dollar hedge fund at a quasi governmental agency. Remember, the Presidential Signing Statement for the Music Modernization Act–signed by Trump 45–specifically designates the MLC board members as inferior officers of the United States. That means on a certain level that they report to the Librarian, a new twist for music business executives. If it comes to a showdown between Trump and the MLC, my money is on Trump. So there’s that.

Time will tell. But one thing is certain: The DOJ lawyer coming in to supervise the entire situation is unlikely to care whether he’ll ever have lunch in that town again.

Can Streaming Price Segmentation Avoid the Malthusian Trap?

When you see artists and songwriters getting starved out of the music business while at the same time fighting over scraps from streaming, that’s unusual. When you see more and more labels caring almost as much about acquiring ever more catalogs rather than helping artists in developing long-term careers, that’s unusual. Why is this becoming the norm? Could it be we are in a “Malthusian trap”?

Remember “nice price” CDs? The budget bin? Top line, mid price, budget price points? That’s called “price differentiation” or “price segmentation.” It’s common in pretty much any consumer good. The idea is that people will pay more for stuff they really want and less money for the nice to haves–the 10 second MBA, buy low sell high. Pretty much any consumer good except–of course–streaming music. One big difference between streaming and physical records is that with streaming, the retailer controls both the wholesale price and the retail price. Want to bet that Wal Mart would just love that model? That should explain why so many artists and especially songwriters are gasping for air. And it should explain why so many are suffering from the streaming pandemic.

Price segmentation in streaming music could be an effective way to avoid the economic concept of the “Malthusian trap.” Simply put, the Malthusian trap occurs when demand for resources outpaces the supply of available resources. This is most likely to happen when buyers with cash exploit sellers who want that cash by using price fixing and market allocation. Like the big pool method of manipulating wholesale prices.

Adopting a more sophisticated approach required by segmentation could allow the music business to move away from the “big pool” program of price controls that has been adopted by every streaming service for both songs and sound recordings.  Notice that blowing up the big pool has nothing to do with a compulsory license.

Remember that the “big pool” method allocates a “pie” which is roughly 50ish% of a defined revenue pool calculated each month for the sound recording and about 14ish% of a slightly different revenue pool for the song. Those two “pies” are then divided up based on market share or said another way, popularity. I say “defined revenue” because it is a negotiated number, not all revenue. Want to bet that defined revenue is less than actual revenue? Sure as there’s gambling at Rick’s. So there’s nothing inherent in the pie, and if you wanted to bet that share price and market valuation is not included in that defined revenue, you’d be a winner.

That “big pool” formula is calculated every month (call that Time X or “Tx”) which is essentially:

(Defined Revenue x [Your Total Streams ÷ All Streams]) = Your Revenue @Tx

and then

Your Revenue ÷ Your Streams = Wholesale Price Per Stream

There are a few bells and whistles to this calculation, but it’s easy to see why this method of price fixing is attractive to the streaming services–it’s just that it’s killing the artists and songwriters stuck in the Malthusian trap. It’s also easy to see that unless Your Total Streams are increasing at a greater rate of increase than the increase in All Streams at T1, T2, T3 and so on, or if the Defined Revenue is not increasing at a greater rate than All Streams, then whoever gets the cash called “Your Revenue” is getting screwed blued and tattooed. Why?

Because they cannot control the wholesale price. That sets into motion the big pool downward spiral and that’ downward spiral can also be called the Malthusian Trap in honor of the 18th Century economist Thomas Malthus who you’ve probably have never heard of.

The Malthusian Trap and Faux Democratization of the Denominator

The “Malthusian Trap” occurs in streaming when wholesale prices determined by the “big pool” method of price caps is overtaken by the services’ open invitation for the supposed “democratization of the denominator.” That surge in tracks uploaded to music streaming services is sometimes estimated at 120,000 per day. (I doubt that it’s exactly that number but let’s go with it on the assumption that whatever the correct number is, it’s a lot compared to what a single artist or even a single label would put into commerce.). You are not uploading 120,000 tracks per day. I doubt that the biggest labels are uploading 120,000 tracks per day and they are definitely not uploading 43,800,000 tracks per year. Granted, those tracks are not all streaming, maybe 25% never get played at all.

But that still means that the only number in the big pool formula that is increasing essentially exponentially is the denominator. And when you consider that streaming revenues are growing less than 10% or so annually, the result of the big pool formula is steadily declining. High school algebra, right?

This faux “democratization” uses artists as human shields to put control of wholesale pricing squarely in the hands of streaming services due to wholesale price caps on both the sound recording and song payouts.

When the growth of a service’s sound recording offering outpaces available revenues, the “big pool” method effectively transfers control over wholesale prices from rights holders to services and causes diminishing returns for both labels and publishers. Regardless of the terms of any one artist’s record deal or the convoluted compulsory mechanical royalty for songwriters, these diminishing returns will hit artists, producers and songwriters because returns are diminished to the labels and publishers, particularly on a per-artist basis.  Particular deals may make the decline even more or less pronounced, but the race to the bottom is baked into the model. High school algebra.

By introducing a more dynamic and differentiated pricing segmentation model, rights holders could regain control over their own wholesale prices, streaming services might better align revenue payouts with actual usage and consumer preferences. We could all potentially avoid the scenario where a fixed revenue pool gets stretched too thin across an ever-expanding catalog.

It must also be said that because performance on Spotify is closely tied–so to speak–to other commerce such as talent buyers for live shows that constantly check how a new artist has performed on Spotify before giving them a show date, a relatively simple economic decision becomes complex. A demonetized artist may be economically indifferent to continuing to support Spotify by driving fans to the platform, but removing themselves from Spotify may hurt them in booking live shows. So the big pool needs to get blown up for yet another valid, if not actionable, reason.

Blowing Up the Compulsory?

On the songwriter side, there is a sense that what we really need to do is blow up the compulsory license particularly given the reaming songwriters are taking from Spotify’s exploitation of the “bundle” loophole that has foolishly been in the Copyright Royalty Board’s regulations for many, many years.  But even so, I suggest that the path dependence of 100-plus years of reliance by a wide variety of music users on the U.S. compulsory mechanical license is unlikely to get “blown up” and abandoned by lawmakers.  But what may get “blown up” is the hated “big pool” royalty payable under that compulsory.  It may turn out that it’s the big pool that is the culprit, not the compulsory license. (And by the way–be careful what you wish for with all this “blowing up” of the compulsory. You may really not want who comes next.)

Why are we still suffering under this ancient regime? Unfortunately, when the handful of people who forced through Title I of the Music Modernization Act got done with it in 2018, they made bad choices.  For example, they had a golden opportunity to do something simple like shorten the rate period from five years to a realistic duration that more closely matches the term of direct license agreements.  It’s simply bizarre to use a five year term during a contemporary era marked by relatively high inflation when rates during the 1988-2004 period were adjusted every two years. 

They also had a chance to choose between perpetuating the DMV-style model of licensing administration in favor of creating an Apple Store-style model and they went for “more DMV please” like carp on bait.  And here we are, more screwed than ever.  Gee thanks, thought leaders!

Understanding the Malthusian Trap in Streaming

In the current “big pool” model, royalties are divided among artists based on the proportion of streams their music receives relative to the total number of streams on the platform. Songwriters are paid in a similar version of the “big pool.”  This system leads to diminishing payouts as the catalog expands and the user base grows, since:

  1. The total revenue pool remains relatively static due to slowing streaming growth and frozen subscription prices, while the denominator (the total number of streams) grows larger;

2. The more content added to the platform, the less valuable each individual stream becomes (regardless of particular artist deals); and

3. Artists or songwriters with fewer streams get demonetized by Spotify or are paid but fall outside the mainstream struggle to receive meaningful payouts.

The “Malthusian trap,” in this context means there is an imbalanced relationship between increasing content and relatively static revenue pools. That imbalance results in declining payouts over time for artists and songwriters. This especially true for those creators whose total streams (the numerator in the ratio) are relatively constant or declining due to falling off in fan engagement for whatever reason (including bands that break up or artists who pass away).

In other words, the big pool’s fixed cap on aggregated streaming prices creates its own scarcity despite the infinite shelf space of a streaming service. (See Chris Anderson’s rather tarnished “long tail” theory that still reigns supreme at streaming services which demonstrates once again there is no free lunch.)

“Malthusian” refers to the sometimes controversial ideas of Thomas Robert Malthus (1766–1834) the British economist, scholar, and demographer, best known for his theories on population growth and its relationship to resources, particularly food. His most influential work is “An Essay on the Principle of Population” (1798), where he argued that populations tend to grow exponentially, while food production grows at a much slower, linear rate.

This mismatch, according to Malthus, would eventually lead to overpopulation and resource scarcity, resulting in widespread poverty, famine, and social instability.  Malthus called this the “surplus population” or what the AI accelerationists call “useless eaters.” Surplus population leads to famine just like streaming leads to Discovery Mode and demonetization.  Mr. Malthus has a fairly gloomy view of the world, so no Spotify stock options for him.  He wouldn’t have his pompoms out as a streaming cheerleader our Thomas, but his ideas are very relevant to the streaming analysis.

Key Concepts of Malthusian Theory:  Also see Malthus critic Charles Dickens (“may I have some more”), England’s response to the Irish potato famine and Gangs of New York.

Exponential Population Growth: Malthus believed that if left unchecked, populations grow exponentially (doubling every 25 years), which would outpace the resources needed to sustain them.  Comparatively, the total number of tracks on Spotify has doubled approximately every four years. (This is like Sergei’s Corollary to Moore’s Law–royalties decline 50% with every two year increase in computing power.)

Limited Food Supply: Malthus argued that food production could only grow at an arithmetic (linear) rate because of the finite land, labor, and capital available to produce it. Over time, the availability of food per capita would diminish just like the per-stream rate on streaming platforms–that’s why Spotify continues to deny a per-stream rate even exists (ludicrous propaganda).  That is, populations tend to increase geometrically (2, 4, 8, 16 …), whereas food reserves grow arithmetically (2, 3, 4, 5 …). I’d say this is like a vast number of under performing recordings lead to competition for the artificially capped revenue under “big pool” and the relatively frozen subscription prices. This helps to explain Daniel Ek’s–very Malthusian–comment that artists need to work harder to keep up which was straight out of Oliver Twist.

The Malthusian Trap: The theory suggests that any improvements in living standards (through better agriculture, technology, or economic progress) would eventually lead to population growth, which would, in turn, bring the standard of living back down to subsistence levels. Essentially, population pressure would cause periodic famines, diseases, or wars–you know, demonetization–that would control population size and maintain balance with available resources. The trap helps to explain why we need to blow up the big pool model and its fixed wholesale prices.

Preventive and Positive Checks: Malthus identified two types of checks on population growth:

Preventive checks: These are voluntary actions people can take to limit population growth, such as delayed marriage and celibacy.  In the streaming analogy, this would occur if Spotify were to limit the number of royalty bearing tracks (like demonetizing under 1,000 streams).

Positive checks: These occur when the population exceeds the capacity for sustenance, leading to famine, disease, and mortality, which ultimately reduce the population.  In the streaming analogy, this occurs when artists or songwriters quit the music business or abandon streaming platforms.  Given the close ties between traction on Spotify and validation for talent buyers, for example, it is unlikely that a working artist could abandon the platform entirely no matter how much it costs them to stay on Spotify–although there are limits.

Can Price Segmentation Address the Malthusian Trap in Streaming?

Price segmentation allows streaming platforms to differentiate pricing based on different user segments, content types, or usage behaviors, which can provide several key benefits to avoid the Malthusian trap. We’ll see if the thought leaders have some other suggestions–that I cynically (I admit) think are most likely to be continuing to put bandaids on the status quo.

Fired for Cause:  @RepFitzgerald Asks for Conditional Redesignation of the MLC

U.S. Representative Scott Fitzgerald joined in the MLC review currently underway and sent a letter to Register of Copyrights Shira Perlmutter on August 29 regarding operational and performance issues relating to the MLC.  The letter was in the context of the five year review for “redesignation” of The MLC, Inc. as the mechanical licensing collective.  (That may be confusing because of the choice of “The MLC” as the name of the operational entity that the government permits to run the mechanical licensing collective.  The main difference is that The MLC, Inc. is an entity that is “designated” or appointed to operationalize the statutory body.  The MLC, Inc. can be replaced.  The mechanical licensing collective (lower case) is the statutory body created by Title I of the Music Modernization Act) and it lasts as long as the MMA is not repealed or modified. Unlikely, but we live in hope.)

I would say that songwriters probably don’t have anything more important to do today in their business beyond reading and understanding Rep. Fitzgerald’s excellent letter.

Rep. Fitzgerald’s letter is important because he proposes that the MLC, Inc. be given a conditional redesignation, not an outright redesignation.  In a nutshell, that is because Rep. Fitzgerald raises many…let’s just say “issues”…that he would like to see fixed before committing to another five years for The MLC, Inc.  As a member of the House Judiciary Subcommittee on Courts, Intellectual Property, and the Internet, Rep. Fitzgerald’s point of view on this subject must be given added gravitas.

In case you’re not following along at home, the Copyright Office is currently conducting an operational and performance review of The MLC, Inc. to determine if it is deserving of being given another five years to operate the mechanical licensing collective.  (See Periodic Review of the Mechanical Licensing Collective and the Digital Licensee Coordinator (Docket 2024-1), available at https://www.copyright.gov/rulemaking/mma-designations/2024/.)

The redesignation process may not be quickly resolved.  It is important to realize that the Copyright Office is not obligated to redesignate The MLC, Inc. by any particular deadline or at all.  It is easy to understand that any redesignation might be contingent on The MLC, Inc. fixing certain…issues…because the redesignation rulemaking is itself an operational and performance review.  It is also easy to understand that the Copyright Office might need to bring in some technical and operational assistance in order to diligence its statutory review obligations.  This could take a while.

Let’s consider the broad strokes of Rep. Fitzgerald’s letter.

Budget Transparency

Rep. Fitzgerald is concerned with a lack of candor and transparency in The MLC, Inc.’s annual report among other things. If you’ve read the MLC’s annual reports, you may agree with me that the reports are long on cheerleading and short on financial facts.  It’s like The MLC, Inc. thought they were answering the question “How can you tolerate your own awesomeness?”   That question is not on the list.  Rep. Fitzgerald says “Unfortunately, the current annual report lacks key data necessary to examine the MLC’s ability to execute these authorities and functions.”  He then goes on to make recommendations for greater transparency in future annual reports.

I agree with Rep. Fitzgerald that these are all important points.  I disagree with him slightly about the timing of this disclosure.  These important disclosures need not be prospective–they could be both prospective and retroactive. I see no reason at all why The MLC, Inc. cannot be required to revise all of its four annual reports filed to date (https://www.themlc.com/governance) in line with this expanded criteria.  I am just guessing, but the kind of detail that Rep. Fitzgerald is focused on are really just data that any business would accumulate or require in the normal course of prudently operating its business.  That suggests to me that there is no additional work required in bringing The MLC, Inc. into compliance; it’s just a matter of disclosure.

There is nothing proprietary about that disclosure and there is no reason to keep secrets about how you handle other people’s money.  It is important to recognize that The MLC, Inc. only handles other people’s money.  It has no revenue because all of the money under its management comes from either royalties that belong to copyright owners or operating capital paid by the services that use the blanket license.  It should not be overlooked that the services rely on the MLC and it has a duty to everyone to properly handle the funds. The MLC, Inc. also operates at the pleasure of the government, so it should not be heard to be too precious about information flow, particularly information related to its own operational performance. Those duties flow in many directions.

Board Neutrality

The board composition of the mechanical licensing collective (and therefore The MLC, Inc.) is set by Congress in Title I.  It should come as no surprise to anyone that the major publishers and their lobbyists who created Title I wrote themselves a winning hand directly into the statute itself.  (And FYI, there is gambling at Rick’s American Café, too.)  As Rep. Fitzgerald says:  

Of the 14 voting members, ten are comprised of music publishers and four are songwriters. Publishers were given a majority of seats in order to assist with the collective’s primary task of matching and distributing royalties. However, the MMA did not provide this allocation in order to convert the MLC into an extension of the music publishers.

I would argue with him about that, too, because I believe that’s exactly what the MMA was intended to do by those who drafted it who also dictated who controlled the pen.  This is a rotten system and it was obviously on its way to putrefaction before the ink was dry.

For context, Section 8 of the Clayton Act, one of our principal antitrust laws, prohibits interlocking boards on competitor corporations.  I’m not saying that The MLC, Inc. has a Section 8 problem–yet–but rather that interlocking boards is a disfavored arrangement by way of understanding Rep. Fitzgerald’s issue with The MLC, Inc.’s form of governance:

Per the MMA, the MLC is required to maintain an independent board of directors. However, what we’ve seen since establishing the collective is anything but independent. For example, in both 2023 and 2024, all ten publishers represented by the voting members on the MLC Board of Directors were also members of the NMPA’s board.  This not only raises questions about the MLC’s ability to act as a “fair” administrator of the blanket license but, more importantly, raises concerns that the MLC is using its expenditures to advance arguments indistinguishable from those of the music publishers-including, at times, arguments contrary to the positions of songwriters and the digital streamers.

Said another way, Rep. Fitzgerald is concerned that The MLC, Inc. is acting very much like HFA did when it was owned by the NMPA.  That would be HFA, the principal vendor of The MLC, Inc. (and that dividing line is blurry, too).

It is important to realize that the gravamen of Rep. Fitzgerald’s complaint (as I understand it) is not solely with the statute, it is with the decisions about how to interpret the statute taken by The MLC, Inc. and not so far countermanded by the Copyright Office in its oversight role.  That’s the best news I’ve had all day.  This conflict and competition issue is easily solved by voluntary action which could be taken immediately (with or without changing the board composition).  In fact, given the sensitivity that large or dominant corporations have about such things, I’m kind of surprised that they walked right into that one.  The devil may be in the details, but God is in the little things.

Investment Policy

Rep. Fitzgerald is also concerned about The MLC, Inc.’s “investment policy.”  Readers will recall that I have been questioning both the provenance and wisdom of The MLC, Inc. unilaterally deciding that it can invest the hundreds of millions in the black box in the open market.  I personally cannot find any authority for such a momentous action in the statute or any regulation.  Rep. Fitzgerald also raises questions about the “investment policy”:

Further, questions remain regarding the MLC’s investment policy by which it may invest royalty and assessment funds. The MLC’s Investment Policy Statement provides little insight into how those funds are invested, their market risk, the revenue generated from those investments, and the percentage of revenue (minus fees) transferred to the copyright owner upon distribution of royalties. I would urge the Copyright Office to require more transparency into these investments as a condition of redesignation.

It should be obvious that The MLC, Inc.’s “investment policy” has taken on a renewed seriousness and can no longer be dodged.

Black Box

It should go without saying that fair distribution of unmatched funds starts with paying the right people.  Not “connect to collect” or “play your part” or any other sloganeering.  Tracking them down. Like orphan works, The MLC, Inc. needs to take active measures to find the people to whom they owe money, not wait for the people who don’t know they are owed to find out that they haven’t been paid.  

Although there are some reasonable boundaries on a cost/benefit analysis of just how much to spend on tracking down people owed small sums, it is important to realize that the extraordinary benefits conferred on digital services by the Music Modernization Act, safe harbors and all, justifies higher expectations of those same services in finding the people they owe money.  The MLC, Inc. is uniquely different than its counterparts in other countries for this reason.

I tried to raise the need for increased vigilance at the MLC during a Copyright Office roundtable on the MMA. I was startled that the then-head of DiMA (since moved on) had the brass to condescend to me as if he had ever paid a royalty or rendered a royalty statement.  I was pointing out that the MLC was different than any other collecting society in the world because the licensees pay the operating costs and received significant legal benefits in return. Those legal benefits took away songwriters’ fundamental rights to protect their interests through enforcing justifiable infringement actions which is not true in other countries.

In countries where the operating cost of their collecting society is deducted from royalties, it is far more appropriate for that society to consider a more restrictive cost/benefit analysis when expending resources to track down the songwriters they owe. This is particularly true when no black box writer is granting nonmonetary consideration like a safe harbor whether they know it or not.

I got an earful from this person about how the services weren’t an open checkbook to track down people they owed money to (try that argument when failing to comply with Know Your Customer laws).  Grocers know more about ham sandwiches than digital services know about copyright owners. The general tone was that I should be grateful to Big Daddy and be more careful how I spend my lunch money. And yes I do resent this paternalistic response which I’m sorry to say was not challenged by the Copyright Office lawyer presiding who shortly thereafter went to work for Spotify.  Nobody ever asked for an open check.  I just asked that they make a greater effort than the effort that got Spotify sued a number of times resulting in over $50 million in settlements, a generous accommodation in my view. If anyone should be grateful, it is the services who should be grateful, not the songwriters.

And yet here we are again in the same place.  Except this time the services have a safe harbor against the entire world which I believe has value greater than the operating costs of the MLC.  I’d be perfectly happy to go back to the way it was before the services got everything they wanted and then some in Title I of the MMA, but I bet I won’t get any takers on that idea.

Instead, I have to congratulate Rep. Fitzgerald for truly excellent work product in his letter and for framing the issue exactly as it should be posed.  Failing to fix these major problems should result in no redesignation—fired for cause.

The Intention of Justice:  In Which The MLC Loses its Way on a Copyright Adventure

ARTHUR

Let’s get back to justice…what is justice?  What is the intention of justice?  The intention of justice is to see that the guilty people are proven guilty and that the innocent are freed.  Simple, isn’t it?  Only it’s not that simple.

From And Justice for All, screenplay written by Valerie Curtin and Barry Levinson

Something very important happened at the MLC on July 9:  The Copyright Office overruled the MLC on the position the MLC (and, in fairness, the NMPA) took on who was entitled to post-termination mechanical royalties under the statutory blanket license.  What’s important about the ruling is not just that the Copyright Office ruled that the MLC’s announced position was “incorrect”—it is that it corrected the MLC’s position that was in direct contravention of prior Copyright Office guidance.  (If this is all news to you, you can get up to speed with this helpful post about the episode on the Copyright Office website or read John Barker’s excellent comment in the rulemaking.)

“Guidance” is a kind way to put it, because the Copyright Office has statutory oversight for the MLC.  That means that on subjects yet to be well defined in a post-Loper world (the Supreme Court decision that reversed “Chevron deference”), I think it’s worth asking whether the Copyright Office is going to need to get more involved with the operations of the MLC.  Alternatively, Congress may have to amend Title I of the Music Modernization Act to fill in the blanks.  Either way, the Copyright Office’s termination ruling is yet another example of why I keep saying that the MLC is a quasi-governmental organization that is, in a way, neither fish nor fowl.  It is both a private organization and a government agency somewhat like the Tennessee Valley Authority.  Whatever it is ultimately ruled to be, it is not like the Harry Fox Agency which in my view has labored for decades under the misapprehension that its decisions carry the effect of law.  Shocking, I know.  But whether it’s the MLC or HFA, when they decide not to pay your money unless you sue them, it may as well be the law.

The MLC’s failure to follow the Copyright Office guidance is not a minor thing.  This obstreperousness has led to significant overpayments to pre-termination copyright owners (who may not even realize they were getting screwed).  This behavior by the MLC is what the British call “bolshy”, a wonderful word describing one who is uncooperative, recalcitrant, or truculent according to the Oxford Dictionary of Modern Slang.  The word is a pejorative adjective derived from Bolshevik.  “Bolshy” invokes lawlessness.

In a strange coincidence, the two most prominent public commenters supporting the MLC’s bolshy position on post-termination payments were the MLC itself and the NMPA, which holds a nonvoting board seat on the MLC’s board of directors.  This stick-togetherness is very reminiscent of what it was like dealing with HFA when the NMPA owned it.  It was hard to tell where one started and the other stopped just like it is now.  (I have often said that a nonvoting board seat is very much like a “board observer” appointed by investors in a startup to essentially spy on the company’s board of directors.  I question why the MLC even needs nonvoting board seats at all given the largely interlocking boards, aside from the obvious answer that the nonvoters have those seats because the lobbyists wrote themselves into Title I of the MMA—you know, the famous “spirit of the MMA”.)

Having said that, the height of bolshiness is captured in this quotation (89 FR 58586 (July 9, 2024)) from the Copyright Office ruling about public comments which the Office had requested (at 56588):

The only commenter to question the Office’s authority was NMPA, which offered various arguments for why the Office lacks authority to issue this [post-termination] rule. None are persuasive. [Ouch.]

NMPA first argued that the Office has no authority under section 702 of the Copyright Act or the MMA to promulgate rules that involve substantive questions of copyright law. This is clearly incorrect. [Double ouch.]

The Office ‘‘has statutory authority to issue regulations necessary to administer the Copyright Act’’ and ‘‘to interpret the Copyright Act.’’  As the [Copyright Office notice of proposed rulemaking] detailed, ‘‘[t]he Office’s authority to interpret [the Copyright Act]  in the context of statutory licenses in particular has long been recognized.’’

Well, no kidding.

What concerns me today is that wherever it originated, the net effect of the MLC’s clearly erroneous and misguided position on termination payments is like so many other “policies” of the MLC:  The gloomy result always seems to be they don’t pay the right person or don’t pay anyone at all in a self-created dispute that so far has proven virtually impossible to undo without action by the Copyright Office (which has other and perhaps better things to do, frankly).  The Copyright Office, publishers and songwriters then have to burn cycles correcting the mistake.  

In the case of the termination issue, the MLC managed to do both: They either paid the wrong person or they held the money.  That’s a pretty neat trick, a feat of financial gymnastics for which there should be an Olympic category.  Or at least a flavor of self-licking ice cream.

The reason the net effect is of concern is that this adventure in copyright has led to a massive screwup in payments illustrating what we call the legal maxim of fubar fugazi snafu.  And no one will be fired.  In fact, we don’t even know which person is responsible for taking the position in the first place.  Somebody did, somebody screwed up, and somebody should be held accountable.

Mr. Barker crystalized this issue in his comment on the Copyright Office termination rulemaking, which I call to your attention (emphasis added):

I do have a concern related to the current matter at hand, which translates to a long-term uneasiness which I believe is appropriate to bring up as part of these comments. That concern is, how did the MLC’s proposed policies [on statutory termination payments] come in to being in the first place? 

The Copyright Office makes clear in its statements in the Proposed Rules publication that “…the MLC adopted a dispute policy concerning termination that does not follow the Office’s rulemaking guidance.”, and that the policy “…decline(d) to heed the Office’s warning…”. Given that the Office observed that “[t]he accurate distribution of royalties under the blanket license to copyright owners is a core objective of the MLC”, it is a bit alarming that the MLC’s proposed policies got published in the first place. 

I am personally only able to come up with two reasons why this occurred. Either the MLC board did not fully understand the impact on termination owners and the future administration of those royalties, or the MLC board DID realize the importance, and were intentional with their guidelines, despite the Copyright Office’s warnings. 

Both conclusions are disturbing, and I believe need to be addressed.

Mr. Barker is more gentlemanly about it than I am, and I freely admit that I have no doubt failed the MLC in courtesy.  I do have a tendency to greet only my brothers, the gospel of Matthew notwithstanding.  Yet it irks me to no end that no one has been held accountable for this debacle and the tremendous productivity cost (and loss) of having to fix it.  Was the MLC’s failed quest to impose its will on society covered by the Administrative Assessment?  If so, why?  If not, who paid for it?  And we should call the episode by its name—it is a debacle, albeit a highly illustrative one. 

But we must address this issue soon and address it unambiguously.  The tendency of bureaucracy is always to grow and the tendency of non-profit organizations is always to seek power as a metric in the absence of for-profit revenue.  Often there are too many people in the organization who are involved in decision-making so that responsibility is too scattered.  

When something goes wrong as it inevitably does, no one ever gets blamed, no one ever gets fired, and it’s very hard to hold any one person accountable because everything is too diffused.  Instead of accepting that inevitable result and trying to narrow accountability down to one person so that an organization is manageable and functioning, the reflex response is often to throw more resources at the problem when more resources, aka money, is obviously not the solution.  The MLC already has more money than they know what to do with thanks to the cornucopia of cash from the Administrative Assessment.  That deep pocket has certainly not led to peace in the valley.

Someone needs to get their arms around this issue and introduce accountability into the process.  That is either the Copyright Office acting in its oversight role, the blanket license users acting in their paymaster role through the DLC, or a future litigant who just gets so fed up with the whole thing that they start suing everyone in sight.   

Saint Thomas Aquinas wrote in Summa Theologica that a just war requires a just cause, a rightful intention and the authority of the sovereign (Summa, Second Part of the Second Part, Question 40).  So it is with litigation.  We have a tendency to dismiss litigation as wasteful or unnecessary with a jerk of the knee, yet that is overbroad and actually wrong.  In some cases the right of the people to sue to enforce their rights is productive, necessary, inevitable and—hopefully—in furtherance of a just cause like its historical antecedents in trial by combat.  

It is also entirely in keeping with our Constitution.  The just lawsuit allows the judiciary to right a wrong when other branches of government fail to act, or as James Madison wrote in Federalist 10, so the government by “…its several constituent parts may…be the means of keeping each other in their proper places.”  

That’s a lesson the MLC, Inc. had to learn the hard way.  Let’s not do that again, shall we not?

Chronology: The Week in Review: The Consensus for Conditional Approval of The MLC, Inc. by the @CopyrightOffice

I am pleased to see that there is a consensus against more happy talk among commenters in The MLC, Inc.’s five year review of its operations at the Copyright Office. The consensus is an effort to actually fix the MLC’s data defects, rogue lawmaking and failure to pay “hundreds of millions of dollars” in black box royalties.

The Two Arguments for Conditional Approval

There is a significant group, and sometimes from unexpected corners, who fall into two broad camps: One camp is approve The MLC, Inc. with post-approval conditions that may lead to being disapproved if not accomplished until the next five year review rolls around.

The other camp, which is the one I’m in if you’re interested, is to spend some time now getting very specific about crucial improvements The MLC, Inc. needs to put into effect and payments they need to make. This would be accomplished by bringing in advisory groups of publishing experts, especially from the independent community, roundtables, other customary tools for public consultations, but the redesignation approval would occur only after The MLC, Inc. accomplished these goals.

Either way, the consensus is for conditions if not the timing. I’m not going to argue for one or the other today, but I have some thoughts about why delayed approval is more likely to accomplish the goals to make things better in the least disruptive way.

Remember, once The MLC, Inc. is approved, or “redsignated,” then all leverage to force change is lost. Putting operations-based obligations on The MLC, Inc. to be responsive to their members before they get the valuable approval preserves leverage and will force change one way or another, The reward for successfully accomplishing these goals is getting approved for another term (or the balance of their five year review). Noah built the Ark before the rain.

What if we fired them?

I’m actually pleased to see the consensus for conditional approval. Simply firing The MLC, Inc. would be disruptive (and they know it), mostly because the Copyright Office hasn’t gotten around to requiring that a succession plan be in place so that firing the MLC would not be disruptive.

The simple solution to this pickle is for the Copyright Office to make any redesignation conditioned upon certain fixes being accomplished on an aggressive time frame. I say aggressive because they’ve had five years to think about this; it shoudln’t take long to at least implement some fixes. But if we don’t make it conditional the MLC will lack the incentive to actually fix the problems.

Conditional Approval

I have to say I was encouraged by the number of commenters who said that The MLC, Inc. needs some very definite performance goals. Many commenters said that those goals needed to be met in order for The MLC, Inc. to get approved for another five years until the next quinquennial review. I’m not quite sure how you approve them for another five years with performance goals unless you are really saying what some commenters came right out and said: Any approval should be conditional.

I think that means that the Copyright Office needs a plan with two broad elements: One, the plan identifies specific performance goals, and then two, establishes a performance timeline that The MLC, Inc. must meet in order for this current “redesignation” to become final.

That “conditional redesignation” would incentivize The MLC, Inc. to actually accomplish specific tasks. The timeline will likely vary based on the particular task concerned, but impliedly would be less than five years. There’s a very good reason to make the approval conditional; there’s just too much money involved. Other people’s money.

The Black Box

Every comment I read brings up the black box. Commenters raised different complaints about how The MLC, Inc. is managing or not managing the matching that is required for the black box distribution contemplated by Congress, but they all were pretty freaked out about how big it is, how little we know about it, and the fact that the board of The MLC, Inc. is deeply conflicted because the lobbyists drafted an eventual market share distribution. Strangely enough, there’s every possibility that the market share distribution will happen, or could happen, right after the redesignation. Also known as losing on purpose in a fixed fight.

There’s an easy fix for that one–don’t do the market share distribution, maybe ever.

The harsh but near certain fact is if there is a market share distribution of the black box, the MLC (and everyone involved) will be sued. It almost doesn’t matter how clean it is. So why do it at all? The MLC is supposed to set an example to the world, right? (And we know how much the world loves it when Americans say that kind of thing.) What if we said that the market share distribution was just bloodlust by the lobbyists salivating over a really big poker pot? On reflection, it should be put aside particularly because Congress may not have been told how big the black box really was if anyone knew at the time. Ahem.

The Interest Penalty

This actually goes hand in hand with another interpretation of the black box provisions of Title I which requires the payment of compound interest for black box money to be paid by The MLC, Inc. to the true copyright owner. That compound interest accrues at the “federal short term rate” in effect from time to time (that rate is adjusted monthly and is currently 5.01%). MLC’s interest obligation accrues in an account set up for the true copyright owner’s benefit, not for the recipients of the market share distribution.

Interest runs from the time the unmatched money is received by the MLC until it is matched and paid. There could easily be several different interest rates in effect if the unmatched royalties stay in the black box for months or particularly years. This concept is elaborated in a comment by the Artist Rights Institute.

Title I requires this “penalty” the same way that it requires the statutory late fee which itself has been the subject of much negotiation. It is important to note that the word “penalty” does not appear in Section 115, but both the interest rate and the late fee are obviously “penalties” in plain English. You don’t have to call it a thing a penalty in order for it to be a penalty. It doesn’t stop being a penalty just because the statute doesn’t define it as one, just like a large furry animal with big teeth, big claws, a loud roar and really bad breath who wants to eat you stops being a bear just because it doesn’t have a sign around its neck saying “BEAR”. Particularly when the furry animal has you by throat.

Align the Incentives

I have to imagine that a penalty of compound interest would incentivize both the MLC and the licensees who pay its bills to clear that black box right quick. If a third party is paying the statutory interest penalty which is how it is now according to MLC CEO Kris Ahrend’s testimony to Congress (under oath), then there’s really no incentive for the MLC to pick up the pace on matching and there’s even less incentive for the licensees to make them do it.

It makes sense that the MLC is to maintain an account for each copyright owner (or maybe for each unmatched song since the copyright owner is not matched), so it only makes sense that these accounts and compound interest would be maintained on the ledger of the MLC. It would be pretty dumb to just lump all the money into one account and run compound interest on the whole thing that would have to be disaggregated every time a song is matched. Assuming matching was the object of the exercise.

Plus, there’s nothing in Title I that says that black box money has to be put in a bank account that accrues interest so that the MLC doesn’t have to pay this penalty for being slow. Again, the word “bank” does not appear in Section 115. It definitely doesn’t say a federally insured bank account, a bank in the Federal Reserve system, or the like–because the statute does not require a bank.

Even so, I have to believe that if you want to an insurance company and said I will bring you the “hundreds of millions of dollars” Mr. Ahrend refers to if you write me a policy that will cover my interest expense and insure the corpus, somebody would take that business. If they can write derivatives contracts for fluctuations in natural gas futures, I bet they could write that policy or my name’s not Jeffrey Skilling.

William of Ockham Gets Into the Act

What makes a lot more sense and is a whole lot simpler is that Congress wanted to incentivize the MLC to match and pay black box royalties quickly. Congress established the compound interest penalty to add jet fuel to that call and response cycle.

That penalty is part of the normal costs of operating the MLC therefore should be paid as part of the administrative assessment, i.e., by the services themselves. If the MLC sits on the money too long, the services can refuse to cover the interest costs beyond that point and the MLC can then pass the hat to the board members who allowed that to happen.

So everyone has a good incentive to clean out the black box. Brilliant lawmaking. I don’t think that’s such a bad deal for the services since they are the ones who sat on the money in the first place that produced the initial hundreds of millions of dollars for the black box. They got everything else they wanted in the MMA, why object to this little detail? Let’s try to hold down the hypocrisy, shall we?

There may be some arguments about that interpretation, but here’s what Congress definitely did not do and about which there should definitely not be an argument. Congress did not authorize the MLC to use the black box money as an investment portfolio. Nowhere in Title I is the MLC authorized to start an investment policy and become a “control person” of mutual funds. Which they have done.

That investment policy also raises the question of who gets the upside and who bears the downside risk. If there’s a downturn, who makes the corpus whole? And, of course, when the ultimate market share distribution occurs, who gets the trading profits? Who gets the compound interest? Surely the smart people thought of this as part of their investment policy.

The Key Takeaway

You may disagree with the Institute’s analysis about what is and isn’t a penalty, and you may disagree about the thing of the conditions on approval, but I think that there is broad agreement that there needs to be a discussion about forcing The MLC, Inc. to do a better job. I bet if you asked, the Congress clearly did not see the Copyright Office’s role as handing out participation trophies or pats on the head. And that should not be the community’s goal, either. This whole thing was cooked up by the lobbyists and they were not interested in any help. That obviously crashed and burned and now we need to help each other to save songwriters today and in future generations. If not us, then who; if not now, then when; if not here, then where?

Who Will Get to the Bottom of The Hundreds of Millions of Black Box Money at MLC?

One of the most common questions we get from songwriters about the MLC concerns the gigantic level of “unmatched funds” that have been sitting in the MLC’s accounts since February 2021.  Are they really just waiting until The MLC, Inc. gets redesignated and then distributes hundreds of millions on a market share basis like the lobbyists drafted into the MMA?  

Not My Monkey

Nobody can believe that the MLC can’t manage to pay out several hundred million dollars of streaming mechanical royalties for over three years so far.  (Resulting in the MLC holding $804,555,579 in stocks as of the end of 2022 on its tax return, Part X, line 11.) The proverbial monkey with a dart board could have paid more songwriters in three years.  Face it—doesn’t it just sound illegal?  In my experience, when something sounds or feels illegal, it probably is.

What’s lacking here is a champion to extract the songwriters’ money.  Clearly the largely unelected smart people in charge could have done something about it by now if they wanted to, but they haven’t.  It’s looking more and more like nobody cares or at least nobody wants to do anything about it.  There is profit in delay.

Or maybe nobody is taking responsibility because there’s nobody to complain to.  Or is there? What if such a champion exists?  What if there were no more waiting?  What if there were someone who could bring the real heat to the situation?

Let’s explore one potentially overlooked angle—a federal agency called the Office of the Inspector General.  Who can bring in the OIG?  Who has jurisdiction?  I think someone does and this is the primary reason why the MLC is different from HFA.

Does The Inspector General Have MLC Jurisdiction?

Who has jurisdiction over the MLC (aside from its severely conflicted board of directors which is not setting the world on fire to pump the hundreds of millions of black box money back into the songwriter economy).  The Music Modernization Act says that the mechanical licensing collective operates at the pleasure of the Congress under the oversight of the U.S. Copyright Office and the OIG has oversight of the Copyright Office through its oversight of the Library of Congress.

But, hold on, you say.  The MLC, Inc. is a private company and the government typically does not have direct oversight over the operations of a private company.

The key concept there is “operates” and that’s the difference between the statutory concept of a mechanical licensing collective and the actual operational collective which is a real company with real employees and real board members.  Kind of like shadows on the wall of a cave for you Plato fans.  Or the magic 8 ball.

The MLC, Inc. is all caught up with the government.  It exists because the government allows it to, it collects money under the government’s blanket mechanical license, its operating costs are set by the government, and its board members are “inferior officers” of the United States.   Even though The MLC, Inc. is technically a private organization, it is at best a quasi-governmental organization, almost like the Tennessee Valley Authority or the Corporation for Public Broadcasting.  So it seems to me that The MLC, Inc. is a stand-in for the federal government.

But The MLC, Inc. is not the federal government.  When Congress passed the MMA and it charged the Copyright Office with oversight of the MLC.  Unfortunately, Congress does not appear to have appropriated funds for the additional oversight work it imposed on the Office.  

Neither did Congress empower the Office to charge the customary reasonable fees to cover the oversight work Congress mandated.  The Copyright Office has an entire fee schedule for its many services, but not MLC oversight.  

Even though the MLC’s operating costs are controlled by the Copyright Royalty Board and paid by the users of the blanket license through an assessment, this assessment money does not cover the transaction cost of having the Copyright Office fulfill an oversight role.

An oversight role may be ill suited to the historical role of the Copyright Office, a pre-New Deal agency with no direct enforcement powers—and no culture of cracking heads about wasteful spending like sending a contingent to Grammy Week.

In fact, there’s an argument that The MLC, Inc. should write a check to the taxpayer to offset the additional costs of MLC oversight.  If that hasn’t happened in five years, it’s probably not going to happen.  

Where Does the Inspector General Fit In?

Fortunately, the Copyright Office has a deep bench to draw on at the Office of the Inspector General for the Library of Congress, currently Dr. Glenda B. Arrington.  That kind of necessary detailed oversight is provided through the OIG’s subpoena power, mutual aid relationships with law enforcement partners as well as its own law enforcement powers as an independent agency of the Department of Homeland Security.  Obviously, all of these functions are desirable but none of them are a cultural fit in the Copyright Office or are a realistic resource allocation.

The OIG is better suited to overseeing waste, fraud and abuse at the MLC given that the traditional role of the Copyright Office does not involve confronting the executives of quasi-governmental organizations like the MLC about their operations, nor does it involve parsing through voluminous accounting statements, tracing financial transactions, demanding answers that the MLC does not want to give, and perhaps even making referrals to the Department of Justice to open investigations into potential malfeasance.  

Or demanding that the MLC set a payment schedule to pry loose the damn black box money.

One of the key roles of the OIG is to conduct audits.  A baseline audit of the MLC, its closely held investment policy and open market trading in hundreds of millions in black box funds might be a good place to start.  

It must be said that the first task of the OIG might be to determine whether Congress ever authorized MLC to “invest” the black box funds in the first place.  Congress is usually very specific about authorizing an agency to “invest” other people’s money, particularly when the people doing the investing are also tasked with finding the proper owners and returning that money to them, with interest. 

None of that customary specificity is present with the MLC.

For example, MLC CEO Kris Ahrens told Congress that the simple requirement that the MLC pay interest on “unmatched” funds in its possession (commonly called “black box”) was the basis on which the MLC was investing hundreds of millions in the open market.  This because he assumed the MLC would have to earn enough from trading securities or other investment income to cover their payment obligations.  That obligation is mostly to cover the federal short term interest rate that the MLC is required to pay on black box.

The Ghost of Grammy Week

The MLC has taken the requirement that the MLC pay interest on black box and bootstrapped that mandate to justify investment of the black box in the open market.  That is quite a bootstrap.

An equally plausible explanation would be that the requirement to pay interest on black box is that the interest is a reasonable cost of the collective to be covered by the administrative assessment.  The plain meaning of the statute reflects the intent of the drafters—the interest payment is a penalty to be paid by the MLC for failing to find the owners of the money in the first place, not an excuse to create a relatively secret $800 million hedge fund for the MLC.  

I say relatively secret because The MLC, Inc. has been given the opportunity to inform Congress of how much money they made or lost in the black box quasi-hedge fund, who bears the risk of loss and who profits from trading.  They have not answered these questions.  Perhaps they could answer them to the OIG getting to the bottom of the coverup.

We do not really know the extent of the MLC’s black box holdings, but it presumably would include the hundreds of millions invested under its stewardship in the $1.9 billion Payton Limited Maturity Fund SI (PYLSX). Based on public SEC filings brought to my attention, The MLC, Inc.’s investment in this fund is sufficient to require disclosure by PYLSX as a “Control Person” that owns 25% or more of PYLSX’s $1.9 billion net asset value. PYLSX is required to disclose the MLC as a Control Person in its fundraising materials to the Securities and Exchange Commission (Form N-1A Registration Statement filed February 28, 2023).  This might be a good place to start.

Otherwise, the MLC’s investment policy makes no sense.  The interest payment is a penalty, and the black box is not a profit center.

But you don’t even have to rely on The MLC, Inc.’s quasi governmental status in order for OIG to exert jurisdiction over the MLC.  It is also good to remember that the Presidential Signing Statement for the Music Modernization Act specifically addresses the role of the MLC’s board of directors as “inferior officers” of the United States:

Because the directors [likely both voting and nonvoting] are inferior officers under the Appointments Clause of the Constitution, the Librarian [of Congress] must approve each subsequent selection of a new director. I expect that the Register of Copyrights will work with the collective, once it has been designated, to ensure that the Librarian retains the ultimate authority, as required by the Constitution, to appoint and remove all directors.

The term “inferior officers” refers to those individuals who occupy positions that wield significant authority, but whose work is directed and supervised at some level by others who were appointed by presidential nomination with the advice and consent of the Senate. Therefore, the OIG could likely review the actions of the MLC’s board (voting and nonvoting members) as they would any other inferior offices of the United States in the normal course of the OIG’s activities.

Next Steps for OIG Investigation

How would the OIG at the Library of Congress actually get involved?  In theory, no additional legislation is necessary and in fact the public might be able to use the OIG whistleblower hotline to persuade the IG to get involved without any other inputs.  The process goes something like this:

  1. Receipt of Allegations: The first step in the OIG investigation process is the receipt of allegations. Allegations of fraud, waste, abuse, and other irregularities concerning LOC  programs and operations like the MLC are received from hotline complaints or other communications. 
  2. Preliminary Review: Once an allegation is received, it undergoes a preliminary review to determine if OIG investigative attention is warranted. This involves determining whether the allegation is credible and reasonably detailed (such as providing a copy of the MLC Congressional testimony including Questions for the Record). If the Office is actually bringing the OIG into the matter, this step would likely be collapsed into investigative action.
  3. Investigative Activity: If the preliminary review warrants further investigation, the OIG conducts the investigation through a variety of activities. These include record reviews and document analysis, witness and subject interviews, IG and grand jury subpoenas, search warrants, special techniques such as consensual monitoring and undercover operations, and coordination with other law enforcement agencies, such as the FBI, as appropriate.  That monitoring might include detailed investigation into the $500,000,000 or more in black box funds, much of which is traded on open market transactions like PYLSX.
  4. Investigative Outputs: Upon completing an investigation, reports and other documents may be written for use by the public, senior decision makers and other stakeholders, including U.S. Attorneys and Copyright Office management. Results of OIG’s administrative investigations, such as employee and program integrity cases, are transmitted to officials for appropriate action. 
  5. Monitoring of Results: The OIG monitors the results of those investigations conducted based on OIG referrals to ensure allegations are sufficiently addressed.

So it seems that the Office of the Inspector General is well suited to assisting the Copyright Office by investigating how the MLC is complying with its statutory financial obligations.  In particular, the OIG is ideally positioned to investigate how the MLC is handling the black box and its open market investments that it so far has refused to disclose to Members of Congress at a Congressional hearing as well as in answers to Questions for the Record from Chairman Issa.

Taking Away the Punchbowl: Removing Open Market Investments from the MLC

Whenever someone is holding your money, you kind of want to a lot of about it, don’t you? You want to know how much, where they hold it, and when you get it back, right? Seems reasonable. But not only does the MLC hold what is likely to be hundreds of millions in black box money, they don’t really tell you these attributes, do they?

Plus they tell you that somehow the Music Modernization Act authorizes them to invest your money in the open market to obtain some theoretical government mandated rate of return, yet the Copyright Act says nothing of the kind.

They then refuse to disclose how they intend to invest OPM and who bears any losses and makes any profit from these open market transactions. They don’t just refuse to tell songwriters, because why tell them whose money it is, they don’t just refuse to tell the Copyright Office who is tasked by Congress with overseeing their operations, they also dodge answering questions from the very Congressional committee that oversees the MLC.

The other problem is that the longer the MLC delays in distributing the black box to the correct copyright owners, the more likely it is that the MLC will choose the nuclear option–market share distribution to the copyright owners who are overrepresented on the MLC’s board without regard to copyright ownership. And just to be clear what that means, it means the black box money is going to get shared with people who aren’t entitled to it thus leaving lots of hungry people.

Many people believe that this is exactly the intent and that the market share distribution will happen right after the MLC gets redesigned to operate for another five years at the punchbowl. Why? Because when the market share distribution happens it very likely will tip off an aggressive no-shit backlash against the MLC that would likely argue they can’t be trusted with much of anything at all.

It’s entirely possible that if the MLC has been this secretive about the amount of black box money in its grasp, the lobbyists never came clean with Congress about just how much the services were paying for their retroactive safe harbor in the form of undisclosed and unallocated monies. Remember the big scramble to deny a press report that the black box was over $1 billion? Maybe that press report wasn’t so far off after all.

Fortunately, there is a simple solution. Congress needs to take away any control or decision making about the unallocated black box money from the MLC. This could be as simple as a technical amendment forcing the MLC to act in a transparent manner and disclose the current investments of the black box money, any desposits and withdrawals, and to force distributions to occur when claimed by the correct owner.

The entire concept of a market share distribution should be eliminated from the Copyright Act because it creates a perverse incentive not to find the true owners by the people who benefit from the market share distribution. This moral hazard was obvious from the time the lobbyists drafted Title I of the Music Modernization Act and it should come as no surprise that it has failed miserably.

There’s no time like the present to fix it. Decisions about the black box should be taken away from the MLC and placed far, far outside if its orbit of the network of interlocking boards, consultants, accountants and companies surrounding the MLC and its confederates.

Will Songwriters Wish they had Gotten Inflation Protection on Streaming Mechanicals?

When the dust settled after the last mechanical royalty rate setting we saw the Copyright Royalty Board approving two different settlements for mechanical royalties. The royalty rate for physical mechanicals and permanent downloads get a significant rate increase and the royalty rate for streaming mechanicals got a theoretical rate increase. However, only physical mechanicals and downloads got both a rate increase and a cost of living adjustment (or “inflation protection”). Streaming mechanicals did not get inflation protection–could have but did not.

This means that the same writers on the same song in the same recording will get inflation protection when that song is sold in physical formats (such as the surging vinyl configuration) or downloads, but will not when that song is sold in streaming formats. What is the logic to this? One difference is that record companies are paying on the physical and download side and the lived experience of record companies necessarily puts them closer to songwriters than the services. And the lived experience of streaming companies is…well, breakfast at Buck’s, Hefner level private jets, warmed bidets and beach volleyball courts at home with imported sand. (Although Sergey Brin has a real beach in his Malibu home. Surf’s up in geekville. Maybe he’ll send DiMA to represent him at the Malibu city council meetings if Malibew-du-bumbum is ready for Silicon Valley style lobbying to decide who can surf Sergey’s beach and the color scheme of their boards. Kind of like the Palo Alto Architectural Review Board with a tan.)

The Big Google

We heard that inflation was transitory, which may prove true–or not. Transitory or not, that’s not an argument against treating songwriters equally on two versions of the same mechanical license; rather, it’s a reason why it should be easy to afford if you cared about sustaining songwriters at least as much as investing in ChatGPT to replace them.

However, in one of the great oopsies of the 21st Century, it doesn’t look much like inflation is all that transitory. Based on some of the posts I wrote starting in 2020, I think we can see that inflation is way worse on the items that count for songwriters like “food at home,” rent, utilities and gasoline. Very often the number of Americans working a job is used to counter the lived experience of the high number of people who believe the economy is tanking. But what about that jobs report? More jobs equals good times, yes? There’s something weird about the math of the jobs report which should make you wonder about whether that’s such a great argument.

If I still have your attention after the “math” word, there are two standard surveys of the economy used to measure jobs that measure different components of the jobs created in a given measurement period. These data are the “Establishment Payroll Survey” which measures the total number of jobs in the U.S. economy. That’s the number most people refer to with the “jobs report” you hear so much about. (More formally titled the “Current Establishment Statistics (Establishment Survey).”)

There’s another number called the “Household Survey” that measures the total number of jobs per household (more formally titled the Current Population Survey).

Note that the Establishment survey measures all jobs; the Household survey measures jobs per household. If you had two or three jobs, the Household survey would count you as “employed”; the Establishment survey would count the number of jobs you had. Now note that there is currently 2.7 million job difference between the two. Why?

I’m not really sure, but it would appear that there are more jobs than households. That difference may occur from time to time, but it’s quite a big difference at the moment and seems to be a trend that’s confirmed by another statistic: the surge in part-time jobs as shown in this chart:

So what’s missing is how many jobs that are counted in the Establishment survey are held by any one or two household members in the Household survey. If you were to draw the conclusion that every job in the Establishment survey is a full time job held by the primary source of support in a household and that when the Establishment number is rising things are looking up, that may be a leap unsupported by evidence. That may be one of the things you’d want to know if you were trying to predict how well the government’s songwriter royalties would hold their value over the five year rate period.

The sharp increase since June in the number of part time workers may suggest that more people are working multiple jobs and not that more people are working. In fact, the total number of full time workers seems to have declined by a bit over the same period.

That’s not to say that inflation protection is not a serious requirement of everyone who relies on the government for their livelihood. While the inflation rate has declined a bit recently, possibly due to the Federal Reserve abandoning its zero interest rate policy, it is still significant. In my view, nothing in the employment report suggests otherwise and continues to highlight the importance of songwriters being accorded the same inflation protection on streaming as they are on physical and downloads.

Just because the physical rate is paid by the record companies and the streaming rate is paid by the richest corporations in history does not excuse the distinction. Each should be protected equally.

Applying a Cost of Living Adjustment to Streaming Mechanicals

You are no doubt aware that the Copyright Royalty Judges handed down a final rule adopting the settlement covering streaming mechanicals reached by the major publishers and the richest and most dominant corporations in the history of Planet Earth: Apple, Amazon, Google, Pandora and Spotify. There are many who are dissatisfied with the negotiated rate, no doubt. There are many who are disappointed that the Judges perpetuated the mind-numbing complexity of the royalty calculation methodology (which probably costs more to account on a per-stream basis than the payable royalty).

That’s all true, but is a byproduct of the discriminatory practices frozen in time at the CRB, a libertarian hell-scape preserved in amber. As if taking a trip to Jurassic World (or at least 2009) wasn’t bad enough, the Judges refused even to place a toe onto the arc of the moral universe as they just did in the Subpart B rate setting in the same proceeding (i.e., the Phonorecords IV rates that abandoned the frozen mechanical and adopted an annual cost of living adjustment for physical and permanent download configurations).

I discuss this in more detail in a post on MusicTechPolicy in which I question whether a hidebound adoption of rates that fail to apply a COLA equally and treat likes alike in the same proceeding is lawful, much less good policy. While the Judges focus on giving the negotiating parties, aka the rich people, what they want and ignore the notorious unfairness of the Copyright Royalty Board whose rulings apply to all songwriters in the world who ever lived or may ever live regardless of representation, I argue that applying the same COLA calculation to streaming as to Subpart B configurations solves the problem. This post will lay out a simple method of implementing a COLA for streaming.

The policy goal would be to apply the COLA formula to streaming. Because the streaming formula is so unduly complex, it’s easy to understand the resistance to adding still another step. Remember that the greater than/lesser of monthly calculation is a series of steps that gets to a per-play rate of sorts. All of the greater of/lesser than calculations have been fought and salivated over by dozens of lawyers (literally) so changing any one of them is probably not productive and in my mind is not necessary to give effect to the COLA. Remember that in the history of the government’s mechanical rate, the COLA was applied to a rate as an uplift, not as a way to calculate a rate. The point of a COLA is always to preserve the value of the government’s rate and recognizes that the songwriter will not have a chance to revisit the rate for five year tranches and a lot can happen in five years.

The easiest way to apply a COLA to streaming is to derive the per-play rate given the current formula and then uplift it with a COLA. The Judges already have a COLA based on CPI-U . The Judges need only apply the COLA as a legal modification to the streaming mechanical and accept the base line rates in the negotiated settlement. Otherwise, the exact same songs with the exact same songwriters for the exact same recording in the exact same proceeding will have a COLA when exploited by record companies and none when exploited by the rich people. This result just seems arbitrary. The labels having shown the way to a fair result should be followed by the DSPs.

We raised this approach in a Phonorecords IV comment I filed for David Lowery, Helienne Lindvall and Blake Morgan:

Applying the COLA to Section 115 may actually have a simple solution. The Judges already have a COLA formula. That formula can simply be applied as a step (5) in 37 CFR §385.21(b). This way the negotiated settlement terms are not re- opened.

Adding a COLA uplift to the applicable royalty calculation is simple. First, determine the applicable payable royalty for the accounting period concerned under the negotiated rates. Then apply the COLA formula derived by the Judges as an uplift to the payable royalties as a last step in the royalty calculation. The COLA could be calculated either annually or monthly although monthly seems more appropriate and accurate.

The uplifted amount (after any uplifted overtime adjustment to plays) would then be reflected on the applicable Copyright Owner’s royalty statement as the payable royalty for that accounting period.

This seems like a simple solution that brings the streaming mechanical out of Jurassic World and into the Era of the Songwriter.