Jay Wolstenholme – Principal at Adkrest – July 2026 – jwolstenholme@adkrest.com

How IBOR/ABOR/Agents Are Advancing to Handle Total Portfolio Allocation (TPA)

For twenty years, the Investment Book of Record (IBOR) and embedded ABOR (Accounting Book of Record) has been treated as bedrock — and rightly so. Before it existed, buy-side portfolio data lived in pieces: scattered across front-office trading systems, middle-office reconciliation desks, and back-office accounting platforms that rarely told the same story at the same moment. IBOR strived to fix that. It gave the industry a path to a single, authoritative, real-time "golden copy" of every trade, position, and transaction, flowing cleanly from the front office through to settlement. It deserves its reputation as one of the most consequential pieces of infrastructure in institutional investing to date.

But now that public, private, and alternative assets have all come into the portfolio, IBOR/ABOR has had to evolve to aggregate, normalize, and process all investment assets to produce the necessary Total Portfolio View.

The Investment Portfolio Has Changed. IBOR/ABOR's Core Principles Haven't.

When the dominant portfolio was publicly traded equities and fixed income — long-only, valued daily off exchange prices — a vertically aligned IBOR was enough. The investment universe was essentially vertical. Assets flowed through a defined processing chain: front office, middle office, back office. A single system, or a stack of layered ones, could capture that chain end to end.

That world is history. Private equity, private credit, infrastructure, direct lending, hedge fund allocations, derivatives overlays, collateral programs, and externally managed mandates now sit shoulder to shoulder with traditional liquid assets. Each brings its own data conventions, valuation cadences, counterparty structures, and operational footprint.

Strategic Asset Allocation (SAA), for all its discipline, institutionalized silos. It defined target weights by asset class and managed each mandate against its own bucket. So the book of record got built the same way: vertically, one silo at a time. The result is structural drift across these silos. We can see each asset class with precision, but the total portfolio almost not at all. In the era of Total Portfolio Allocation (TPA), where every holding needs to be aggregated into one view for both transactional processing and risk and performance analytics, the siloed constraints are no longer a quirk — they are a risk and an analytical blind spot.

The siloed architecture stacked an Execution Management System on top of an Order Management System, wired to a Portfolio Management System, which was then handed off to accounting. Every layer and silo demands a reconciliation before anyone can assemble a coherent picture. Each reconciliation buys latency, operational risk, and cost. For institutions operating at scale on thin margins, that is not a trade-off worth defending anymore. It's a tax we've simply gotten used to paying, and it's leaking basis points that margins can no longer tolerate.

TPBOR Is the Next Evolution of IBOR/ABOR — a Complete Architecture That Can Finally Be Built

The Total Portfolio Book of Record (TPBOR) does not displace IBOR/ABOR. It elevates it. Where IBOR (ABOR included) captured the transactional truth of individual asset portfolios — the SAA worldview — TPBOR captures the enterprise truth: the complete, unified picture of every holding across every asset class, regardless of who manages it, how it's valued, or through what structure it's held. It takes the logic of the golden copy and extends it horizontally across the full inventory — public and private, liquid and illiquid, internally run and externally delegated, across the dozen-plus external asset managers a serious asset owner might now use.

TPBOR is not merely a data-architecture upgrade. It is a shift in how investment decisions get framed — from asset-class management to enterprise-level portfolio stewardship.

As SAA has evolved into TPA, the issues across assets have grown complex — each asset class has its own demands, and normalization is complicated. Public holdings carry daily, often intraday, NAVs. Private equity and infrastructure are valued quarterly, with interim valuations coming from models rather than markets. A real Total Portfolio Book of Record must hold both asset timing rhythms without distorting one against the other — consistent standards and governance that let you compare exposures across instruments that simply don't report on the same clock.

Agentic AI With Orchestrated Agents Is What Finally Makes This Achievable

A Total Portfolio Book of Record contains the right data architecture. But architecture alone doesn't act. Aggregating, normalizing, stress-testing, and acting upon total portfolio data — across asset classes on different timelines, in different systems, under different risk models — demands a new kind of operational intelligence. That intelligence has arrived in the form of agentic AI, and it may be the difference between TPBOR as a concept and TPBOR as a realistic working operating model. Legacy connection and aggregation code can no longer plug the dam.

Credit still needs to be given to the current generation of AI tools that came before agentic AI agents. Generative LLM models, analytical dashboards, and AI copilots have been useful and progressive. They synthesized information, surfaced insights, and sped up human analysis.

But they are passive. They wait to be asked. They answer questions instead of pursuing outcomes. We told ourselves that keeping a human in the loop at every step was prudent, but keeping a human in the loop to always trigger the next action can also be a bottleneck.

Agentic AI elevates the workflow, automating repetitive processes and freeing asset managers, asset owners, and traders for high-value decision making. Agentic AI perceives its environment through live feeds and system integrations, reasons over structured and unstructured information, plans multi-step workflows with conditional logic, and executes within defined governance boundaries. Critically, it can move along both axes at once — horizontally across asset classes and vertically up and down the transactional lifecycle chain — in a single coordinated workflow.

Orchestrated agents pull current positions across every asset class, check exposures against compliance and regulatory limits and the Investment Policy Statement, run factor-based risk-performance simulations that blend public market data with private asset valuations, identify rebalancing opportunities spanning liquid and illiquid holdings, draft the compliance and covenant documentation, and route the package for human review. In today's operating model, each of those steps is a hand-off and a reconciliation across separate teams and systems. In an agentic model, they are staged in a unified, governed workflow.

The real power shows up when specialized agents are orchestrated in unison. A liquidity agent, a risk-decomposition agent, a compliance-monitoring agent, a corporate-actions agent, a macro-factor agent — all running at once, each drawing on the TPBOR data layer, each feeding a coordinated matrix of portfolio-level recommendations, rebalancings, and risk-performance analysis. The Investment Policy Statement, the risk budget, and the regulatory constraints become the operating charter under which every agent acts. What you get isn't a pile of disconnected analytics — it's a single, enterprise-level recommendation grounded in the full complexity of the total portfolio.

To the objection that autonomy means losing control, the opposite is true here. Every autonomous action generates a structured audit trail capturing not just what happened, but why: which data sources were queried, which constraints applied, which intermediate results shaped the output. That is more transparency than most reconciliation-heavy manual operational processes can produce today, not less. In addition, an abundance of detailed data is produced for post-trade analytics, uncovering bottlenecks and inefficiencies.

Without leveraging AI agents, firms face continued false/positive manual workflow intervention, difficulty integrating unstructured data, and distraction of portfolio managers, traders, and operational personnel from high-cost, high-risk issues — to name just a few. Agentic AI is the assistant everyone has always needed: one that never leaves and is always looking for more repetitive tasks to take on.

The Destination Is the Total Portfolio View Realized in Production Reality

Where this leads is a genuine Total Portfolio View: a real-time or near-real-time picture of every holding across the enterprise — public and private, liquid and illiquid, directly held and externally delegated, syndicated loans — read through consistent risk, factor, and liquidity lenses, and made actionable by an agentic layer that turns the view into coordinated decisions. TPBOR is the data and function backbone; agentic AI is the operating intelligence. The payoff is the ability to respond to macro shocks dynamically, rebalance across the entire inventory with precision, and manage factor exposure at the enterprise level rather than one silo at a time.

This is where Strategic Asset Allocation (SAA) finally yields to Total Portfolio Allocation (TPA) as the governing framework. It changes the question from "which manager holds what percentage in which asset class?" to "what is our true economic risk exposure, across every instrument and structure we hold, and does it match our beliefs and our tolerance at the total portfolio level?" That is the difference between administering a collection of portfolios and running a single enterprise balance sheet with intent.

So here is the bottom line. The institutions building this infrastructure now aren't modernizing their plumbing — they're constructing a moat. They are buying the capacity to act decisively, transparently, and in concert across the whole balance sheet, at the speed markets move, clients expect, and regulators increasingly demand. TPA is the investment scope; TPBOR, leveraging the best of agentic AI, is the technical architecture to construct a production Total Portfolio View.