Summary

  • Resolution 201302.163 establishes three things: AFRINIC’s Board made the decision, the Secretariat proposed and submitted the material, and the approved object was “Investment Policies and Guidelines.” It does not establish the exact February date, the vote, the policy clauses, any holding, any transaction, implementation, compliance or return.
  • The closest disclosed context is consequential but bounded. At the end of 2012 AFRINIC reported USD 1,168,021 in total reserves, USD 2,636,032 in membership-fee income, USD 25,264 in interest income, USD 115,728 in exchange gain and a USD 263,091 deficit. Those figures make treasury discipline a member-accountability question; they do not identify an investment portfolio or prove why the policy was adopted.
  • A defensible investment framework for a thin private coordinator would put continuity, liquidity and principal protection before bounded return. Members should be able to see its purpose, risk envelope, liquidity floor, authority map, reporting rhythm, exception logic and aggregate compliance, while operationally sensitive account and dealing details remain confidential.
  • AFRINIC is a private technical bookkeeper and coordinator. Its authority over its own organisational funds is ordinary corporate stewardship. It creates no sovereign, regulatory, police, punitive, confiscatory or adjudicative power over networks, members or Internet number resources.

A resolution with a large blank behind it

The public act is brief enough to be clear and incomplete at the same time. Under February 2013, AFRINIC’s Board register records Resolution 201302.163: the Board approved the Investment Policies and Guidelines proposed and submitted by the Secretariat. The wording identifies the decision maker, the submitting body and the object approved. It confirms that the Board did not merely note a discussion or invite future work. It approved something presented to it. That is the firm institutional core of the record.

The same entry does not reproduce the approved document or link to it. It supplies no exact day in February. It names no mover or seconder, lists no attendees, records no quorum finding and gives no vote tally, abstention, recusal, conflict declaration, dissent, committee recommendation, adviser or explanation of the Board’s discussion. None of those silences proves that a vote, quorum check, disclosure or review did not occur. They establish only that the surviving public entry does not allow a reader to examine them. Evidentiary discipline requires keeping that distinction intact.

The missing text is even more important than the missing meeting detail. Resolution 201302.163 states no permitted asset class and no prohibited instrument. It provides no credit-quality requirement, maturity limit, currency rule, concentration ceiling, counterparty limit, liquidity floor, custody arrangement, benchmark, loss tolerance, rebalancing method or exception procedure. It does not say who could instruct a bank, value an asset, select a counterparty, approve a departure or report a breach. Those are not obscure accessories to an investment policy.

They are the controls through which the title “Investment Policies and Guidelines” would acquire operational meaning.

The nearest disclosed financial scale explains why the blank cannot be treated as trivial. AFRINIC’s 2012 annual report recorded USD 1,168,021 in year-end total reserves or total assets attributable to members, compared with USD 1,431,112 in 2011. The 2012 figure is not a portfolio balance, not a cash balance, not a segregated reserve and not proof of an amount invested under the February decision. It is the closest disclosed pre-resolution measure of accumulated institutional value. Used within that boundary, it shows that treasury governance concerned value material to the continuity of a narrow member-funded organisation.

The composition of disclosed income sharpens the accountability question. The annual report recorded USD 2,636,032 in membership-fee income and USD 121,931 in grants. Membership fees were therefore about 95.58 per cent of those two reported income lines. This is a calculation about the composition of two disclosed figures, not a tracing exercise: it cannot prove that every dollar of reserves came from fees. It certainly cannot convert a contractual fee paid to a private coordinator into a sovereign tax. It does show why members had a direct stake in whether organisational money was protected, liquid and handled under bounded delegation.

Treasury effects were already visible in the accounts. The 2012 summary included USD 25,264 in interest income, USD 115,728 in exchange gain and USD 861 in other costs, producing reported total other income and expenses of USD 140,132. These figures do not identify deposits, currencies, instruments, counterparties or the risk accepted to produce the result. They do not reveal a portfolio. They simply demonstrate that interest, currency exposure or other treasury effects could reach the financial statements even before the public register recorded approval of the investment framework.

The same report showed a USD 263,091 deficit and an equal decline in total reserves. That preceding condition makes continuity and capital preservation material, but it is not evidence of motive. The public record does not say the deficit caused the February policy, that the policy was meant to repair it or that any investment outcome later offset it. The chronology permits a careful sentence—AFRINIC approved an investment framework after a year in which it reported a deficit and lower reserves—but it does not permit a causal one.

This is the central asymmetry of Resolution 201302.163. The act of adoption is visible, while the substance, execution and outcome are not. A terse resolution can prove that a governance object exists. It cannot prove that the object was well designed, that staff followed it, that investments were made, that the organisation gained or lost money, that a limit was breached, that an auditor objected, that members challenged it or that the Board later amended it. The honest investigation must resist both reflexive trust and reflexive accusation.

Continuity money, not a speculative mandate

An investment policy at a registry should begin by stating what the money is for. AFRINIC performs a thin coordination function. Its continuity depends on being able to meet near-term operating needs, sustain staff and systems, and preserve the recordkeeping service through pressure. Funds accumulated from a predominantly fee-supported model are therefore continuity money before they are an opportunity to seek return. The legitimate ordering is continuity first, liquidity second, principal protection third and only then a bounded return compatible with the first three obligations.

That priority order is a governance test, not a recovered clause from the unseen policy. Nothing in the public entry tells us how the Board ordered its objectives. Yet the order matters because each investment choice can trade one objective against another. A longer maturity may offer a different return while making cash less available. Exposure to a currency can alter reported value when exchange rates move. Concentrating deposits with one institution can simplify administration while increasing counterparty dependence. An external manager can add expertise while also adding fees, incentives and another layer of delegated judgment.

The first control is principal protection. For an organisation with a narrow operational purpose, an avoidable loss does more than reduce a balance-sheet line. It can narrow the time available to respond to an interruption, increase pressure on future fees or make ordinary service costs harder to absorb. Principal protection does not mean that every financial value is perfectly static; the disclosed exchange gain itself shows that currency effects can touch reported results.

It means the Board should decide and disclose how much loss risk, if any, is compatible with continuity, rather than letting return seeking silently define the institution’s risk appetite.

Liquidity is a separate control. A financial claim may appear valuable while being unavailable when payroll, systems or other near-term needs must be met. A policy suitable for continuity would therefore distinguish operating cash from money that can remain untouched for longer. It would define a minimum liquid floor, the horizon it covers and the process for restoring that floor after use. Again, these are the features against which a policy should be assessed. Resolution 201302.163 does not tell us whether AFRINIC adopted them.

Duration controls turn that liquidity objective into measurable boundaries. The longer money is committed, the more its availability can diverge from operating needs. A board should know the maturity distribution it is permitting and who can extend it. A member-facing summary need not reveal the exact date of every deposit or transaction. It can disclose duration bands, ceilings and aggregate compliance. Without that architecture, readers cannot tell whether “investment” meant disciplined cash management or a broader mandate.

Credit and counterparty risk concern who owes the money back and how much organisational value depends on that promise. A prudent design can set quality thresholds, prohibit unacceptable exposures, cap the amount entrusted to any single counterparty and require periodic review. It can also define how rapidly a deteriorating exposure must be addressed. The public resolution gives none of this. No criticism of a named bank or any presumed provider follows, because the record identifies no bank, custodian, manager, adviser or investment counterparty under the February policy.

Currency risk is particularly visible as a category because the annual report separately recorded an exchange gain. That figure proves an accounting effect, not a strategy. It does not establish which currencies AFRINIC held, why it held them, whether an exposure was deliberate, or whether the later policy constrained it. A public policy skeleton could nonetheless explain the acceptable relationship between currencies used for operating obligations and currencies in which reserves may be held. It could report exposures in aggregate without providing instructions that would compromise account security or negotiating flexibility.

Concentration is broader than currency. It can arise from placing too much with one institution, in one type of claim, within one maturity band or under one delegated manager. The purpose of a concentration ceiling is not to decorate a policy with numbers. It is to prevent a single failure or judgment from threatening continuity money. A credible public account would state the categories subject to limits and report compliance against them. The February entry does not reveal whether such limits existed, so no later observer should imagine them into the record.

Custody and execution controls determine whether an approved risk envelope survives contact with practice. The policy architecture should say which roles can propose a transaction, approve it, execute it, hold the resulting asset, reconcile the record and report the position. Separating these functions can make errors and unauthorised action easier to detect. Dual authorisation, transaction thresholds and independent reconciliation are examples of relevant controls, but they remain design tests here, not claims about what the Secretariat proposed or what the Board approved.

Fees and valuation deserve equal attention. A seemingly modest return can be reduced by management, custody or transaction costs. A valuation method can determine how a gain, loss or compliance position appears in a report. A board should specify what costs are acceptable, how they are disclosed, how holdings are valued and how valuation uncertainty is handled. Resolution 201302.163 reports none of those terms, while the closed record provides no amount of investment fees, no holdings and no performance measure attributable to the policy.

Finally, exceptions reveal whether limits are actual controls or merely aspirations. A useful framework would state who may approve a departure, for what reason, for how long and with what remedy and retrospective review. It would preserve a breach or exception record even when no public disclosure of a transaction is appropriate. An annual aggregate account could tell members how many exceptions occurred, whether they were resolved and whether the policy changed as a result. There is no evidence here that an exception occurred; the point is that an accountable framework makes exceptions visible without inviting invention.

What the resolution proves—and what it cannot carry

Institutional documents are strongest when read for the act they actually record. AFRINIC’s Board register is authoritative evidence that AFRINIC recorded Resolution 201302.163 and described the Board as approving investment policies and guidelines proposed and submitted by the Secretariat. The register’s institutional vocabulary is not self-validating beyond that. The word “approved” does not demonstrate that the underlying analysis was rigorous. The words “policies and guidelines” do not demonstrate that the rules were complete. The fact of Board action does not demonstrate implementation or performance.

The proposer-to-approver path does offer a governance question. The Secretariat prepared or at least submitted the material, and the Board approved it. A responsible framework would clarify how management’s operational knowledge informed the proposal, how directors tested the proposal, which roles were delegated after approval and how results returned to the Board for oversight. But the entry does not disclose a paper trail, adviser input, risk analysis or committee recommendation. The most that can be said is that a Secretariat-to-Board proposal path is visible at the level of the resolution sentence.

Nor does the record establish the quality of collective decision-making. There may have been extensive consideration behind the public line; there may have been little. There may have been unanimous support, divided views, abstention or recusal; none is disclosed. Silence about conflicts is not proof of a conflict, just as silence about dissent is not proof of consensus. Good accountability would publish enough of the approval chain to let members distinguish considered authorisation from a bare institutional assertion, but historical uncertainty must remain uncertainty.

The same restraint applies to implementation. No source here identifies a first transaction, an implementation date, a holding, a mandate given to a manager, a custody account, a compliance certificate, a benchmark, a gain, a loss, an impairment or an amendment. The 2012 interest and exchange figures precede the resolution and cannot be attributed to it. The total reserves figure cannot be relabelled as the amount governed by it. There is no established member challenge, audit finding or policy breach. Any story that fills those blanks would cease to be an investigation and become fiction.

There is also no basis for allegations of illegality, corruption, theft, self-dealing, misappropriation, concealment, recklessness or bad faith. A missing public attachment is an accountability gap, not proof of wrongdoing. The gap matters because it prevents verification and makes institutional memory fragile. But the remedy for unverified design is to request the design and its compliance evidence, not to invent a scandal that the evidence does not support.

This disciplined boundary permits a stronger conclusion than either accusation or deference. AFRINIC created an internal treasury-control object at Board level. Members cannot determine from the public record what risks it bounded or whether it worked. Therefore the appropriate demand is not that AFRINIC disclose every operational detail, nor that members accept the title of a policy as proof. The demand is for the public architecture by which stewardship can be judged.

The adjacent April controls are not the missing policy

The 2013 Board register records separate action in April. It says the Board adopted Delegation of Authority 002-2013, named signatories for bank accounts and required Board approval for transactions above USD 100,000. It also says bankers were to receive relevant policies and resolutions. These are adjacent bank-authorisation controls. They help show that the Board was addressing financial authority in the same general period, but they do not disclose the February investment policy.

The chronology tempts a neat reconstruction: a February investment framework, followed by April execution controls. The public record does not make that connection. It does not say the April resolutions implemented, amended or formed part of Resolution 201302.163. It does not identify the February investment policy among the materials sent to bankers. The USD 100,000 threshold belongs to the separate April record and cannot be imported into the unseen February instrument.

Keeping the instruments separate is more than archival caution. Governance often fails in analysis when documents with similar subject matter are fused into a single imagined system. A bank-signatory rule answers who may authorise certain account activity. An investment policy should answer why money is held, what risk may be accepted, how liquidity is preserved and how performance and compliance are monitored. The two may interact in practice, but interaction must be shown, not presumed. Here it remains unknown.

The April measures also demonstrate why publication of an authority map would be useful. Readers can see a transaction threshold in one instrument and approval of a policy in another, yet they cannot tell how the roles fit together. A redacted governance summary could show which decisions belonged to the Board, a committee, the Secretariat, finance staff, a bank, a custodian or an external manager. It could cross-reference separate authorities without publishing account names or dealing instructions. The surviving register does not provide that map.

The private-bookkeeper boundary

AFRINIC is a private technical bookkeeper and coordinator. That is the controlling institutional truth. The registry ledger exists to serve operational reality; it does not create operational reality and it is not a throne. AFRINIC is not a sovereign, government, regulator, police force, prosecutor, punishment authority, confiscator or court. Its corporate decision to govern its own organisational money cannot become authority over the legal or economic fate of networks or Internet number resources.

This boundary is crucial in a treasury story because money can make institutional language sound grander than the underlying mandate. Member fees finance a private coordination service. They are not taxation. Reserves support the organisation’s capacity to deliver that service. They are not a public treasury. Board approval is corporate stewardship within contract and company law. It is not an exercise of public fiscal power. The discipline expected of the Board follows from entrustment and continuity, not from sovereignty.

The distinction narrows both the institution’s claim and the members’ demand. AFRINIC may legitimately define how its funds are held and who can act for the company. Members may legitimately demand economy, liquidity, preservation, clear delegation and evidence of compliance because their fees sustain the service. Neither proposition turns the registry into a ruler of the Internet. A sound investment policy protects the conditions under which a replaceable coordinator performs a limited job; it does not expand that job.

This is also why institutional growth, prestige and return seeking cannot justify themselves. The purpose of holding and investing organisational money must remain tied to the thin registry function. Return may reduce future fee pressure or help preserve value, but only within bounds compatible with service continuity. The organisation does not earn a wider mandate merely because it can accumulate reserves or employ financial expertise. Its legitimacy rests on doing the narrow job economically and accountably.

The ledger’s service character changes the meaning of risk. A liquidity failure could impair staffing, systems or registry-service continuity. That operational consequence is serious. Yet seriousness is not a warrant for coercive authority. It is a reason to make controls narrow, auditable and portable, and to preserve the possibility that functions can continue even if an institution performs poorly. Governance repair should enhance replaceability, not use continuity anxiety to place the bookkeeper above the networks it records.

The public account of Resolution 201302.163 should therefore be judged on two axes. The first is stewardship: did the policy protect principal, liquidity and continuity under clear delegation? The second is mandate restraint: did internal financial control remain internal rather than support claims of public authority? The surviving entry cannot answer the first because the policy and performance record are absent. The controlling doctrine answers the second: no internal investment decision can confer sovereign or punitive power.

The strongest case for confidentiality

There is a serious benign case for a short public resolution above a confidential operating document. A responsible Board should not let staff improvise with reserves. The Secretariat may have been best placed to convert operating needs into a workable treasury framework, and Board approval may have imposed professional discipline on management. Publishing account names, credentials, live allocations, transaction timing, dealing instructions or sensitive counterparty terms could create security risks or weaken the organisation’s negotiating position.

That case must be presented in full because transparency is not the indiscriminate release of every financial detail. A treasury policy can contain information whose immediate publication would make execution harder or less safe. Directors also need room to obtain advice and compare commercial terms. If the underlying February document was competent and confidential for such reasons, the terse resolution would sit above a real control framework even though the public entry alone cannot prove it.

Operational confidentiality, however, does not require architectural secrecy. Members do not need passwords, account identifiers or a schedule announcing when a transaction will occur. They do need to know the policy’s purpose and the ordering of continuity, liquidity, principal protection and return. They need the categories of permitted and prohibited risk, the minimum liquidity standard, the boundaries of delegated authority, the frequency of reporting, the logic for exceptions and an aggregate statement of compliance.

The difference is between information that enables interference with execution and information that enables judgment of governance. A counterparty’s confidential commercial term may belong in the first category. A maximum counterparty concentration, expressed as a policy ceiling and reported in aggregate, belongs in the second. A live dealing instruction can remain private. The fact that one person cannot propose, approve, execute and reconcile the same transaction should be public. Account credentials must remain secret. The existence and cadence of independent reconciliation should not.

This distinction also protects the Board. When only the title of a policy is public, later directors cannot easily demonstrate that predecessors designed a defensible system, and members cannot separate competent confidentiality from empty formality. Publishing the governance skeleton creates a stable institutional memory. It permits oversight without inviting outsiders into operations and reduces the space in which both unwarranted suspicion and unwarranted confidence grow.

A useful disclosure could therefore be redacted rather than absent. It might show objectives in priority order, the division between near-term liquidity and longer-horizon protected reserves, permitted asset and currency ranges, credit, duration and concentration ceilings, custody and delegation roles, benchmark and fee rules, exception authority, quarterly aggregate compliance and an annual member review. Sensitive transaction-level and account-level material would remain confidential. This is a counterfactual design, not a description of AFRINIC’s unseen policy.

The benign case consequently strengthens, rather than defeats, the request for disclosure. If the Board approved a capable control framework, a public skeleton should allow it to demonstrate that fact without sacrificing security. If the framework was thin, the same disclosure would allow improvement. The principle is proportionality: protect what must remain operationally confidential, publish what members need to assess entrusted stewardship.

The information members should have been able to test

The first test is purpose. The Board should state why reserves exist and how investment serves the registry’s limited function. A purpose clause should prevent return from becoming an independent institutional ambition. It should explain the horizon against which continuity is measured and the circumstances in which reserves may be used. Resolution 201302.163 tells the public only that policies and guidelines were approved, leaving this basic orientation unknown.

The second test is the risk envelope. Members should be able to see categories of permitted and prohibited exposure, minimum credit quality where relevant, maximum maturity, acceptable currency relationships and concentration ceilings. A public statement can describe these boundaries without naming every instrument or institution. Such disclosure would not show that a particular transaction occurred. It would show what the Board authorised management to consider.

The third test is liquidity. An organisation funded mainly through membership fees must be able to show that foreseeable operational needs are not dependent on selling or unwinding longer-horizon positions at the wrong time. A floor can be expressed as a horizon or proportion, with aggregate compliance reported periodically. The exact floor adopted in 2013, if any, is not in the record. That absence is precisely why the control should be part of the public architecture.

The fourth test is authority. The policy should map proposal, approval, execution, custody, reconciliation, valuation, monitoring and exception roles. It should show which matters the Board retains and which it delegates. It should specify how conflicts are handled and how delegated actors are reviewed. The register identifies the Secretariat as proposer and the Board as approver at adoption, but it gives no post-adoption operating map.

The fifth test is monitoring. Holdings, maturity, currency, risk, costs, benchmark and compliance should return to the Board at a defined cadence. Members need not receive a live position report. They should receive a periodic aggregate account sufficient to tell whether the policy’s objectives and limits are being met. No reporting frequency, benchmark, valuation method or compliance certificate is disclosed for Resolution 201302.163.

The sixth test is exception control. A departure should identify the authoriser, reason, duration, remediation and later review. Aggregate publication can state whether exceptions occurred without disclosing sensitive execution detail. That record helps distinguish a flexible but controlled framework from one in which limits can be waived invisibly. Nothing in the evidence establishes an exception or breach; it establishes only that the public cannot inspect the exception design.

The seventh test is member accountability. An annual review should connect the policy to the limited service members finance. It should explain whether the liquidity purpose remains appropriate, whether aggregate exposures remained within bounds, what fees were incurred and whether the framework needs revision. That review should not recast members as subjects of a public authority. It should treat them as the parties financing and relying on a private coordination service.

Together these tests create a decision-ready standard. They do not assume that one asset, bank, currency or manager is inherently appropriate, and they do not manufacture historical terms. They ask whether the Board made purpose, risk, liquidity, authority, monitoring, exceptions and accountability visible enough to judge. On the surviving public record, adoption passes the existence test. Every substantive test remains unanswered.