[Week 24, 2025] Capital Allocation and Human Resource Problems

Welcome back to Price and Prejudice with a few notes from Week 24 of 2025, connecting headlines to deeper themes in markets and human behavior.

Battle of Business Models

This FT piece explores how Blackstone, KKR, and Apollo – once near-identical leveraged buyout shops – are pursuing radically different strategies as market conditions tighten. Blackstone remains asset-light, collecting fees without balance sheet risk; Apollo has morphed into a lending behemoth through insurance acquisitions; KKR splits the difference with a Berkshire-style permanent capital vehicle alongside its insurance operations.

The strategic divergence mirrors, in some ways, the subtle but meaningful differences among the passive indexing giants. Vanguard is owned by its funds, which are owned by investors. So it’s basically a co-op: it exists to lower costs for its shareholders. BlackRock is a for-profit firm with a massive tech business (Aladdin) and lots of institutional clients. And State Street is more niche – it’s mostly an institutional custodian bank that also runs index funds, especially for pensions.

In both cases, what appeared to be commodity businesses (buy everything, charge low fees) have revealed themselves to be quite different creatures under stress, with management teams making pointed remarks about whose approach will prove more durable when the cycle finally turns.

Replicating Berkshire

The perennial problem in investing (and in life) surfaces again when examining exceptional success: how much reflects replicable strategy versus unrepeatable circumstances? It's usually both, but distinguishing between the two matters if you're trying to learn something useful rather than just marvel at outcomes.

This article from Barron's catalogs six companies attempting the Berkshire playbook: combining insurance float with long-term investing, with results that illuminate this challenge perfectly. Fairfax and Markel have generated impressive returns over decades, suggesting the model has genuine merit, while Greenlight Capital Re trades below its 2007 IPO price. What appears to be a simple formula (collect premiums, invest the float, and repeat) isn't as straightforward as we thought.

AI in Central Banking

The Bank of International Settlements (BIS) is often called the "central bank of central banks," and it has a knack for publishing some of the most thoughtful analysis on emerging financial technologies, often well before these topics hit mainstream policy discussions. This latest bulletin continues that tradition by mapping out two potential futures for artificial intelligence in central banking: (1) "AI copilots" that enhance human capabilities, and (2) "AI agents" that could autonomously handle specific tasks like real-time forecasting or transaction verification.

What's particularly interesting is how the challenge from AI is framed as a human resources problem rather than a technological one. For example, 83% of surveyed central banks report increased complexity in workforce planning, with particular difficulty recruiting in cybersecurity, data science, and AI/ML roles. Thus the implicit tension: central banks need to attract talent from a private sector that can offer significantly higher compensation, while maintaining their traditionally cautious, consensus-driven culture.

This talent competition isn't unique to central banking, of course. Government agencies and academic institutions (!) have long struggled to compete with private sector compensation for technical talent. The traditional solution has been to emphasize mission-driven work, job security, and non-monetary benefits like research freedom or public service impact. But central banking may be uniquely ill-suited to this playbook. The work, however important, lacks the immediate social impact of healthcare or education, while the institutional culture of deliberate consensus-building may frustrate technologists accustomed to rapid iteration and experimentation.

Life as the Sum of Choices

I recently watched Mission: Impossible – The Final Reckoning, and a line stuck with me (and apparently also with some Redditors): “Our lives are the sum of our choices.” The phrasing immediately calls to mind the fundamental valuation principle that prices are the sum of discounted cash flows – in some sense, both attempts to collapse complex realities into clean mathematical relationships.

But the analogy breaks down in an interesting way. In finance, the cash flows themselves aren't really choice variables; they're projections of what a business might generate based on market conditions, competition, and countless external factors. Our choice is binary and simple: do we take on the project or not? But life reverses this structure entirely. We didn't choose to exist – that decision was made for us – but the individual choices that follow are genuinely ours to make. In that sense, we're the asset that's already been deployed, stuck generating our own cash flows through the choices we make, with no external investor able to divest when performance disappoints. And Tom Cruise, at 62 and still doing his own stunts, embodies this notion perfectly.