Ask an artist and the answer is usually “the curator.” Ask a curator and you get a more complicated story involving a committee, a donor, a registrar’s condition report, a storage audit and a trustee who liked something at an art fair. The gap between those two answers is where most misunderstanding about museum collections lives — including the widespread and largely mistaken belief that acquisition is a matter of individual taste.
It is better understood as a procurement process with an aesthetic argument attached, and the aesthetic argument is frequently not the part that fails.
The formal chain
Most public museums of any size follow a version of the same sequence, whatever it is called locally.
A curator proposes. The proposal is a written document, not a conversation: what the work is, its provenance and condition, why it belongs in this collection, what gap it fills, what it cost, what it will cost to keep, and what comparable works have sold for. In many institutions this is three to ten pages and takes weeks to prepare.
Internal review. The registrar checks provenance and title. Conservation assesses condition and long-term care requirements. Collections management asks the unglamorous and often decisive question of where it will physically go. Finance confirms the money exists in the right budget line.
A committee decides. Usually an acquisitions committee of the board, meeting a handful of times a year, composed of trustees — often collectors themselves — sometimes with external advisers and the director. Curatorial staff present; the committee votes.
The board ratifies, and for some institutions a government or ministry sign-off follows, particularly where export licences, national-treasure rules or state ownership are involved.
Two things about this chain are worth noticing. The curator’s expertise enters at the start and is channelled through a document that must persuade non-specialists. And the committee that decides is usually not composed of art historians.
The people with real influence, ranked honestly
The director sets the collecting priorities that determine which proposals are even written. A director who has decided the institution must address a historical absence in its holdings has already decided most of what the next five years of acquisitions will be. This is the most consequential and least visible power in the process.
Curators identify, research, argue and shepherd. Their power is agenda-setting rather than decisive: they choose what to propose, which is substantial, but they rarely choose what is accepted.
The acquisitions committee and board hold the formal authority. Their influence runs well beyond the vote, because curators learn to propose what will pass. A committee with known tastes shapes proposals it never sees.
Donors, who are the single most underestimated force. The majority of works in most encyclopaedic museum collections arrived as gifts or bequests, not purchases. A gift is cheap to accept and expensive to refuse — refusing offends a donor who may be funding a wing — which means the path of least resistance has shaped collections more than any curatorial vision. Donors also attach conditions: that the work be displayed, that it never be sold, that it hang near something else, that it be shown within a certain number of years. Those conditions bind the institution for decades.
Conservation and the registrar hold a quiet veto. A work that cannot be stored safely, cannot be insured, has an unresolved provenance gap, or requires conditions the building cannot provide will not be acquired regardless of how badly the curator wants it. This is where more proposals die than anywhere else.
Dealers and galleries control access and pricing, time offers to coincide with committee calendars, and maintain museum-discount practices that make institutional purchases possible. They also decide which museum gets first refusal on a sought-after work, which is real influence over the shape of public collections.
Government, in systems where it applies: export licensing, acquisition subsidies, national collection strategies, and in some countries direct ministerial approval.
What the committee is actually weighing
Not “is this good.” The questions in the room are more specific, and knowing them explains most outcomes.
Does it fill a documented gap? Collections are assessed against a written collecting policy. A work that strengthens an acknowledged weakness beats a better work that duplicates existing holdings.
Is the provenance clean? Since the 1998 Washington Principles on Nazi-confiscated art and the 1970 UNESCO Convention threshold for antiquities, provenance research has moved from courtesy to prerequisite. A gap in the ownership record between 1933 and 1945, or an antiquity without documented pre-1970 export, will stop an acquisition at most reputable institutions now — and the restitution claims of the past two decades have made boards considerably more risk-averse than they were.
What does it cost to keep? The purchase price is often the smaller number. Storage, climate control, insurance, periodic conservation, and — for time-based, digital and performance work — active maintenance, migration of obsolete formats, and the cost of re-staging. A video work acquired today commits the institution to decades of format migration. This is now a serious constraint on collecting contemporary art.
Can it be shown? Scale is a real limit. Works that will never leave storage are harder to justify, which quietly biases collections toward the exhibitable.
Does it carry risk? Political controversy, contested cultural ownership, materials that degrade, artists with reputational problems, works whose display might require a conversation the institution does not want to have.
Deaccessioning, the mirror image
The reverse decision is more tightly controlled than acquisition and more often contested. Professional codes — ICOM’s, and the American Alliance of Museums’ more specific rules — hold that proceeds from selling collection objects may be used only for acquisition or, under narrower terms, direct collection care, never for operating costs.
The principle was tested publicly during the pandemic, when the AAM temporarily relaxed restrictions and several US museums sold works to fund operations, prompting sanctions and a sustained argument about whether collections are assets or held in trust. The dispute matters here because it exposes what acquisition assumes: that a museum takes on a work permanently, which is why the entry process is as cautious as it is.
What this means for a living artist
The practical conclusion is unglamorous, and it is the opposite of what most artists assume.
You cannot submit work to a museum collection. There is no application. Unsolicited offers of your own work are declined as a matter of policy at essentially every institution, and the reason is not snobbery — accepting work directly from artists creates valuation problems, sets a precedent that generates hundreds of offers, and bypasses the external validation the process relies on.
Works enter through intermediaries. A gallery places work with a museum. A collector buys from a gallery and later gives or bequeaths it. A curator encounters the work in an exhibition, a biennial or a studio visit, and proposes it years later. Almost every contemporary work in a public collection arrived by one of those three routes.
Exhibition history is the currency. Curators propose work they have seen, in contexts their committee will recognise. Showing in a respected non-commercial space does more for eventual institutional interest than any direct approach.
Documentation matters at the point of acquisition. When a curator does write a proposal about your work, the information available determines whether it survives internal review. Materials, installation requirements, condition, provenance of every prior sale, and your own instructions about display and future conservation — if these are not recorded, the conservation assessment becomes a set of unknowns, and unknowns lose votes. Artists who keep serious records are easier to acquire.
Smaller and specialised institutions are more reachable, and often more useful. University museums, regional collections, artist-run archives and foundation collections have shorter chains of decision and explicit mandates to collect emerging or local work. A work in a small public collection has entered the institutional record, and the institutional record is cumulative.
The answer
Nobody decides alone, and the decision is distributed in a way that is neither as arbitrary as artists fear nor as expert as the public assumes.
A curator with scholarly authority proposes. A committee of trustees, many of them collectors rather than scholars, approves. A conservator and a registrar hold a veto on practical and legal grounds. A donor’s preferences, expressed through a gift, bypass much of the argument entirely. And a director, by setting collecting priorities, determines the range of what anyone will propose in the first place.
Which means the honest answer to who decides what enters a museum is: a small group of people, some of them not art historians, applying a written policy under financial and legal constraints, several years after the work was made. The taste of a single individual is almost never the deciding factor — and the parts of the process that most often stop an acquisition are storage, provenance and the cost of long-term care.
