Disintermediation¶
Reroute an economic flow away from a specified incumbent intermediary layer without implying that every intermediation function disappears.
Core Idea¶
Disintermediation is a change in an economic route: a specified incumbent intermediary ceases to carry, or carries less of, an affected flow because that flow moves through another path. The unit of analysis matters. In the 1966 Federal Open Market Committee discussion, holders of negotiable bank certificates of deposit were drawn toward higher-yielding market securities under Regulation Q rate ceilings; the speaker explicitly called the run-off from banks “disintermediation.” In a later content-distribution case, Disney shifted content previously offered through Netflix and Amazon to its own Disney+ service. These do not share the same goods, institutions or trigger. They share a before-and-after bypass of an identifiable incumbent layer.[1][2]
“Direct” is relative, not an assertion that all mediation or exchange functions vanish. The 1966 CD-to-securities shift does not establish that the same savers funded the same borrowers through a perfectly direct link; it establishes reduced claims on banks and greater investor holdings of market paper. Disney's change concerned named digital intermediaries while its linear television arrangements remained. Other actors or systems still handle trading, access, delivery, payment and governance. Research on re-intermediation supplies a further warning that eliminating one institutional intermediary can coexist with new intermediary roles elsewhere.[1][2][3]
The idea is therefore a relative pathway transformation, not a promise of lower cost, better welfare or an intermediary-free economy. Its diagnosis asks which flow left which middle layer, where it went, and which functions were reassigned.
Structural Signature¶
- Prior intermediated flow: identify the claims, funds, goods or content that reached participants through an incumbent layer. Without a prior mediated path, there is no bypass relative to that layer.
- Specified incumbent intermediary: name the bank, distributor, platform or other layer whose position in the affected flow changes. “The middleman” without a reference layer cannot be tested.[1][2]
- Alternative route: a market instrument or supplier-operated channel carries the affected flow without that specified layer. It need not be entirely free of other institutions.
- Observable bypass or displacement: compare the old and new routes. A merely proposed direct channel, or a new channel with no displaced flow, does not yet establish the change.[1][2]
- Residual function allocation: track who now performs access, liquidity, matching, delivery or governance. This is a necessary interpretive check against the false inference that bypassing one intermediary annihilates its tasks, although no single replacement organization is constitutive.[1][3]
Sig role-phrases: prior intermediated flow — specified incumbent intermediary — alternative route — observable bypass or displacement — residual function allocation.
What It Is Not¶
Disintermediation is not synonymous with all trade becoming two-party direct trade. Market securities still have counterparties and trading arrangements; a supplier-operated streaming service still needs technical and commercial support. The claim is always relative to a layer and a flow. It is not automatically reintermediation, which concerns new or repositioned intermediary roles after a bypass; the latter may follow, but need not be asserted for every case.[1][3]
It is also not merely adding a direct channel. Schauerte and colleagues make a specific supplier-level distinction: severing intermediary ties and selling through the supplier's own channel is disintermediation, whereas maintaining the old relationship alongside the new route is supplier encroachment. At the transaction level, some flows could bypass an intermediary even within a dual-channel system, so the analytic level must be declared rather than silently switching definitions.[2]
Nor is it channel conflict. Conflict requires a consequential struggle with incumbents—such as lost margin or retaliation—and can arise even if the old channel persists. A bypassed layer may also disappear without the particular hold-up/retaliation mechanism represented by the live Channel Conflict node. A cause such as a regulation, digital technology or margin is contingent, not constitutive; the observable route change is.
Scope of Application¶
The literal scope is markets and distribution systems with a discernible mediated route. In finance, depositors or investors can shift claims away from deposit-taking institutions toward securities; the classic Regulation Q episode is one historical case. The Federal Reserve speaker's account specifies a rate-ceiling/yield difference and explains why the shift altered bank balance sheets, while carefully treating aggregate-credit effects as a separate matter.[1] The pattern is not limited to that regulation or to the 1960s.
In nonfinancial distribution, a supplier can bypass a retailer, publisher or streaming intermediary for a specified product flow. The Disney+ study offers a source-grounded example and distinguishes the displaced digital channel from continued linear-TV relationships. Its empirical demand effects are contingent on product and consumer circumstances; they are not part of the definition.[2] The 2025 blockchain case study is used only for the separate function-persistence boundary, not as evidence that banking and media have identical intermediary technologies.[3]
Clarity¶
The concept separates where value or claims move from whether the entire system becomes direct. In the 1966 FOMC discussion, a bank CD holder's claim on a bank shrinks or lapses, and a security holding replaces it. Calling that “saver lends directly to the bank's former mortgage borrower” would invent a counterparty relation the source does not establish. The supported claim is the bypass of deposit-taking bank intermediation for the marginal funds under discussion.[1]
Similarly, “Disney disintermediated Netflix/Amazon” is precise only for the affected Disney content and digital channel. It does not mean the firm had no other distributors or that every consumer migrated. The original article explicitly notes continued non-digital intermediary relationships and distinguishes those from the retired digital distribution route.[2]
Manages Complexity¶
A real market path has many layers: deposit-taking, security issuance, exchange, settlement, content licensing, discovery, delivery and payment. Describing all of them as either simply “intermediated” or “direct” loses the specific structural change. Disintermediation compresses the before/after comparison to a named incumbent layer and tracked flow, then obliges the analyst to uncompress the residual functions before claiming an efficiency result.
This avoids two errors at once. First, an incumbent's fall in activity is not automatically the disappearance of the underlying economic demand. The FOMC speaker separated falling bank assets from what he modeled as the supply of credit to the economy.[1] Second, eliminating one gateway does not mean no other gateway forms; Feulner and coauthors document new governance, compliance and technical-integration roles in their distinct blockchain cases.[3]
Abstract Reasoning¶
Begin by drawing the old route and marking the incumbent intermediary. Select a concrete flow—CD balances, a content catalog, a class of orders—rather than an entire industry. Draw the alternative route and ask what evidence shows the selected flow actually moved. Then identify the incentive or technical condition that made the change possible in this case, and follow the incumbent's former functions to whoever now performs them. Finally keep the consequences separate: cost, demand, credit access and welfare require additional evidence.[1][2]
This procedure distinguishes a genuine change from a publicity claim. A manufacturer may advertise a new web store but leave its wholesale relationships and most transactions unchanged; under the supplier-level convention in Schauerte et al. that is encroachment, not full disintermediation. Conversely, a partial withdrawal of a specified content catalog from a digital intermediary is observable disintermediation of that flow even while a separate linear-TV route survives.[2]
Knowledge Transfer¶
The mapping from bank finance to entertainment distribution preserves the relation prior mediated flow → named incumbent layer → alternative route → displaced flow. It does not preserve the objects being exchanged or the exact trigger. A yield ceiling explains the bank case, while strategic control of a supplier-operated service explains the media case. Importing the first cause into the second would be a false analogy.[1][2]
The abstraction transfers within economic channel analysis to other cases only after the same roles are evidenced. The most general graph operation—remove or bypass an intermediate node—may be portable beyond markets, but this named entry is not demonstrated here as a cross-domain prime. Its financial and distribution examples remain two forms of market intermediation, and its consequences depend on institutional details.
Examples¶
1966 bank-CD runoff. In the FOMC memorandum, Governor Mitchell described market yields above Regulation Q ceilings as drawing holders of maturing negotiable CDs from bank claims into short-term securities. His analysis identifies bank disintermediation but does not prove that every affected dollar directly reached the same borrower, nor that aggregate credit must fall by the same amount as bank deposits.[1] Mapped back: prior intermediated flow = investor funds held as negotiable bank CDs; specified incumbent intermediary = deposit-taking banks; alternative route = investor holdings of short-term securities; observable bypass or displacement = CD runoff accompanied by fewer bank claims and more public securities holdings; residual function allocation = security issuance, trading and settlement still occur, without an asserted same-borrower direct link.
Disney's digital distribution change. Schauerte and coauthors study Disney+'s introduction after Disney withdrew content previously made available through digital intermediaries including Netflix and Amazon. The supplier-operated streaming path bypassed those digital layers for the affected content; separate non-digital linear-TV relationships were maintained. The research examines differing demand responses rather than claiming inevitable gains.[2] Mapped back: prior intermediated flow = Disney content supplied through digital streaming intermediaries; specified incumbent intermediary = Netflix/Amazon for that affected digital flow; alternative route = Disney+; observable bypass or displacement = withdrawal from named digital intermediaries and offer through the supplier-operated service; residual function allocation = platform operation and content delivery remain, while non-digital distribution still uses intermediaries.
These cases are unlike in institution, asset, trigger and direction of economic claims. The common identity survives only because the same relative layer-bypass test can be applied precisely to each.
Structural Tensions¶
Bypass of an incumbent versus continued need for its functions. Removing a bank deposit path or a licensed digital distributor can reduce reliance on that institution, but financing access, liquidity, discovery, delivery or governance still have to be provided somewhere. Optimizing simply for fewer named intermediaries can shift work and cost onto users or newly created agents. Diagnostic: After the flow changes route, which actor performs each former intermediary function, and what evidence shows the work is cheaper or better?[1][2][3]
Direct-route control versus incumbent reach and complementary channels. A supplier-operated path can increase control over content and customer access while giving up some benefit of an established distributor's reach; the effects may differ across digital and non-digital incumbents. These poles cannot be maximized just by relabeling every path “direct”: retaining some incumbent relationships may preserve value but changes the scope of the disintermediation claim. Diagnostic: Which exact flow is rerouted, what reach or service did its former intermediary provide, and does the evidence support a net gain rather than merely a new path?[2]
Structural–Framed Character¶
Disintermediation lies toward the framed side of the structural–framed spectrum. The route change has a portable graph-like skeleton, but its named identity is used in market arrangements whose legal and contractual roles define who counts as an intermediary. It carries some evaluative weight in ordinary usage (“cut out the middleman” often suggests efficiency), whereas the analytic definition here is neutral. It depends on human practices of exchange, deposits, distribution and platform access, not just a physical path. Its history and vocabulary arise in economic and marketing institutions. “Bypass” travels widely, but importing “disintermediation” into an unrelated graph is metaphor until a comparable intermediary role and economic flow are established.
Live Indirection is not a strict parent: its definition concerns interposed references that decouple providers and consumers, while this entry removes or diminishes a market actor's route position. A general intermediate-node-bypass abstraction could be a future-prime question, but two economic substrates do not establish it. Its character: a reusable, partly structural but institutionally framed market-path transformation, properly domain-specific at this evidence level.
Structural Core vs. Domain Accent¶
The core is a relative comparison: the same specified category of economic flow formerly used an incumbent intermediary and now uses an alternative path that bypasses that layer. Regulation Q, CDs and securities are financial accents. Streaming catalogs, platform licensing and supplier-operated digital delivery are entertainment-distribution accents. An Internet channel or lower cost may be common but is not necessary. The apparently portable skeleton is bypass of an intermediate node; whether that merits a separate prime is unresolved and should not be inferred from these market cases alone.[1][2]
No current live parent passes the strict test. Channel is a bounded information conduit with capacity/noise commitments, not the genus of institutional market paths. Common-Medium Intermediation installs a hub; Indirection inserts a reference; neither is a kind of layer removal. The live Channel Conflict entry concerns a damaged distribution partner and retaliation, a possible accompaniment but not a necessary part of every reroute. Staging this node unparented preserves those boundaries pending later DAG densification.
Instantiates / Related Primes¶
No strict typed edge is asserted. Indirection and Common-Medium Intermediation are useful contrasts because each explains a positive intermediary structure that may be bypassed, but disintermediation is not a subtype of either. Channel Conflict is a neighboring potential consequence, not a parent or synonym. Credit channel describes one financial setting's transmission mechanism, not the general route change. The current graph proposal therefore has an explicit unparented status rather than a forced lexical relation.
The refusal to attach a parent is provisional. If future curation supplies a verified market-layer-bypass genus, this relation can be reconsidered.
Neighborhood in Abstraction Space¶
Disintermediation sits in a sparse region of the domain-specific corpus (98th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2551 abstractions)
Nearest neighbors
- Supplier Concentration Risk — 0.76
- Measuring network throughput — 0.76
- Sales cannibalization — 0.75
- Sudden Stop (Economics) — 0.75
- Cumulative flow diagram — 0.75
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
- Supplier encroachment or dual distribution: a supplier adds a direct channel while retaining its incumbent intermediary relationship; Schauerte et al. use this as a supplier-level distinction, though individual transactions may still bypass a layer.[2]
- Reintermediation: a new or repositioned middle actor takes on functions after a bypass, a possible later development rather than the same event.[3]
- Channel conflict: harm or retaliation among existing distribution partners, which may occur with no completed removal of an incumbent layer.
- Total removal of intermediary functions: not entailed by reduced use of one institution; financial trading and digital delivery still require coordination.[1][3]
- An automatic efficiency or welfare improvement: bank credit and media demand effects require their own case-specific evidence.[1][2]
References¶
[1] Federal Open Market Committee, “Memorandum of Discussion,” 13 September 1966, Governor George W. Mitchell's statement, printed pp. 56–59. Original contemporaneous record inspected. Its forward-looking and illustrative monetary effects are his policy analysis, not independently verified universal results. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p
[2] Nico Schauerte, Ricarda Schauerte, Maren Becker and Thorsten Hennig-Thurau, “Making new enemies: How suppliers’ digital disintermediation strategy shifts consumers’ use of incumbent offerings,” Journal of the Academy of Marketing Science 52 (2024), 672–694; first published 21 August 2023. Original open-access empirical article inspected, especially abstract, introduction and “Review of disintermediation and the related encroachment concept.” registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p
[3] Simon Feulner, Tobias Guggenberger, Jens-Christian Stoetzer and Nils Urbach, “Beyond disintermediation: A multiple case study of emerging intermediary roles in blockchain applications,” Electronic Markets 35, article 98 (2025), abstract and introduction. Original open-access study inspected only to qualify the claim about persistence/reassignment of intermediary functions; it does not establish the financial or media cases. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h