Perspectives

Prologue

We began writing this article after a spirited discussion over dinner. It was one of those evenings where no one remembers the appetizer, but everyone remembers the argument. The argument centered on the Internal Rate of Return (IRR) calculation, and whether it embeds a reinvestment assumption that distorts private market investment allocations. Some of us left business school believing that IRR carries a reinvestment rate burden; others were convinced that believing this was itself the burden. Somewhere between course and discourse, the seeds of this article were sown.

Executive Summary

In public markets, instant price discovery and time-weighted returns represent standard features. Historically, however, capital formation¹ primarily occurred through private, opaque, relationship-driven transactions, where cash-flow metrics like the internal rate of return (IRR) have long dominated. These cash flow metrics remain the norm in contemporary private markets, particularly within the general partner/limited partner (GP/LP) structure² that governs the institutional private equity and private credit markets.

Our central argument is this: the widespread belief that IRR calculations require reinvestment assumptions is simply incorrect. A persistent myth suggests IRR is only valid if you can reinvest distributions at the same rate of return, which may lead to a systematic under allocation to private markets. This misunderstanding may inadvertently lead investors to dismiss legitimate IRR performance or to resort to other metrics, such as time-weighted returns (TWRs), which were designed for different contexts and come with their own limitations.

Furthermore, high IRRs of individual investments do not necessarily translate into higher TWR at the portfolio level. In this article, we aim to dispel the myth about reinvestment rate assumptions, and to illustrate the differences between IRR and TWR. Understanding these differences is essential for making informed investment decisions — and avoiding costly mistakes.

Instant Price Discovery: The Exception, Not The Rule

Throughout history, real-time price discovery has been remarkably uncommon. Ancient financial systems, from Mesopotamia and Greece to Rome, relied predominantly on privately negotiated arrangements. Medieval and Renaissance financiers in Florence, Venice, and Amsterdam similarly prioritized relationships and trust over market-based transparency, conducting business through personal networks built over generations.

Even in modern times, public market exchanges like NYSE and LSE, though substantial, are dwarfed by the scale of privately held capital, which includes private equity, private debt, real estate, infrastructure, natural resources, family businesses, partnerships, and lending from one neighbor to another.

Why does private financing continue to flourish despite the advent of sophisticated public exchanges and remarkable technological advancements? Several factors explain the persistence of private market dominance.

First, real-time public pricing requires a robust regulatory regime and a significant investment in technology, which are features often unique to developed economies. Second, many assets that change hands in private markets are inherently complex and difficult to value through standardized mechanisms. Consider the challenge of establishing an instant market price for an airport or a professional sports franchise. These assets require expert evaluation, detailed due diligence, and negotiated transactions.

Finally, private entities may prefer limited disclosure to maintain a competitive advantage or to focus on long-term strategic objectives rather than succumb to short-term market pressures that often characterize public company management.

Time-Weighted Returns: An Exception, Not A Rule

As mutual funds gained popularity in the mid-20th century, regulators and industry groups like the CFA institute faced a critical challenge: how to create a standardized way to compare portfolio managers’ investment skill. In public markets, investors rather than portfolio managers control the timing of cash inflows and outflows in public markets. It would be misleading to reward or penalize portfolio managers for timing decisions beyond their control.

Thus, the Global Investment Performance Standards promoted the time-weighted return (TWR) as the default approach for presenting public-market track records. By removing the effects of investor deposits and withdrawals, TWRs create an ideal framework for assessing the performance of publicly traded portfolios like mutual funds, where portfolio managers only control their period-to-period investment decisions, not the capital flows.

However, in most other investment contexts — particularly for those involving irregular or uneven cash flows — TWRs become less relevant than cash-flow-based metrics like the IRR, net present value (NPV), or multiples on invested capital (MOIC). The reason for this is cash flow-based return metrics focus on the actual economic outcome for investors by factoring in the precise timing and amount of each cash flow, both contributions and distributions.

Proponents of TWRs may argue that IRR is too influenced by the timing and sizing of cashflows. This impact is real and must not be ignored, but other metrics can be examined alongside IRRs to create a balanced view of performance. Duration² is one such measure, as it provides important context and is usually calculable using information provided by private capital managers.

Why IRR Does Not Require a Reinvestment Rate Assumption — And Why It Matters

A common misconception pervades some finance literature and business school classrooms: the belief that IRR calculation implicitly assumes the reinvestment of interim distributions at the IRR. However, the mathematical solution to the IRR algorithm is simply the discount rate that equates the present value of future (expected) investment cash flows to zero. This definition of IRR is strictly internal to the specific investment (hence the name) and is independent of subsequent investment decisions. IRR measures what happened within the boundaries of one particular investment — nothing more and nothing less.

The confusion originates from literature that discusses reinvestment rates in the context of total realized returns over extended periods. This is relevant when evaluating an investor's overall performance across multiple investments but does not pertain to the IRR calculation itself, which is a project-specific measure devoid of any reinvestment assumptions.

Many practitioners learned from older finance textbooks that discuss the “reinvestment rate assumption.” These texts sometimes state: “If you want to measure your total realized return over time — including what you do with distributions — the rate at which you redeploy those payouts will affect your overall return.” This statement is undeniably true, but fundamentally different from claiming that “the IRR calculation itself requires you to assume a particular reinvestment rate.”

The IRR is simply a solution to a present-value equation; it does not embed any assumption about what you do with the distributed cash beyond the project’s boundaries. The distinction between measuring individual investment performance and measuring overall investor performance across multiple investments is crucial — confusing them leads to flawed decisions.

In the modern private capital universe, investors sometimes argue that while the IRR of a private fund might reach 12%, they should discount this reported return because they cannot necessarily reinvest distributions at 12%. This conflates the actual performance of the fund (which may truly generate a 12% IRR on deployed capital) with the investor’s overall realized return across multiple investments (which depends on how the investor redeploys the proceeds).

This incorrect assumption can have significant practical impact. Investors may extrapolate these perceived lower fund returns to the asset class level, perhaps using a modified IRR function with a lower reinvestment rate. All else being equal, a lower expected return for a given asset class may lead to a reduced allocation in a strategic asset allocation exercise, potentially costing investors substantial returns over time.

We argue that the fund’s IRR remains valid as a performance metric for that fund’s capital usage. Including a reinvestment assumption necessarily introduces the performance of a second investment into the equation. While this approach may be meaningful from a portfolio standpoint, where one may want to form a holistic view of all investments combined, we should avoid doing so when considering only one investment in isolation.

IRRs and TWRs at the Portfolio Level

If we move beyond the decision of one investment and instead consider the performance measurement of a portfolio of investments, it is worth considering how IRR and TWR relate. One observation is that investors (limited partners) may be graded on their TWR production while they assess managers (general partners) based on IRR production. This begs the question of whether and to what extent these two measures correlate.

We argue that this relationship is not necessarily intuitive and – performance persistence issues aside – IRR may not be the best proxy for TWR, and vice versa. Understanding this distinction becomes crucial for both investment selection and performance evaluation.

Consider two portfolios to illustrate this point: Portfolio A generates a 26% IRR by returning $126 million from an initial $100 million outflow after one year and does not reinvest. Portfolio B returns a modest 8% annually over three years, but capital remains invested the entire time.

Despite Portfolio A's significantly higher IRR, the portfolio's overall TWR is the same as Portfolio B’s due to the latter’s continuous compounding of returns. Portfolio B may in fact be more attractive to a portfolio manager due to its consistent cashflows over a longer time horizon and because it obviates the need to reinvest cashflows. However, this preference does not negate the objectively higher IRR of Portfolio A.

This example demonstrates that IRR and TWR do not necessarily correlate intuitively at the portfolio level, which highlights the distinct roles and interpretations of these metrics. No single perfect metric exists, but an understanding of their behavior can create a balanced view of performance.

Conclusion

Instant price discovery and TWRs are modern constructs that are predominantly observed in specialized public market contexts. Contemporary private equity and private credit, which often feature large, complex, and unique transactions, as well as the global market for private financing more broadly, tend to rely on cash flow-based metrics like IRR, and for good reason.

The myth that IRR inherently assumes interim reinvestment at the same rate of return is just that—a myth. IRR is purely internal and project specific, measuring only what happens within the boundaries of a single investment. Misinterpreting IRR can lead investors to underestimate the true effectiveness of private funds and make suboptimal allocation decisions.

Additionally, recognizing the limited relationship between investment-level IRR and portfolio-level TWR may help investors better evaluate the performance of their portfolios. Investors making allocation decisions should accept IRR without reinvestment assumptions and be mindful of its relationship with TWR, despite what they may have learned in business school.

In a world where private markets play an increasingly important role, getting the measurement right directly impacts financial outcomes for millions. Clarity in performance measurement leads to better investment decisions — a recipe that everyone at the dinner table can agree on.

William P. Kieser, Ph.D., Principal, Quantitative Research Group within Ares Global Client Solutions; Eyal Karsh, Head of Private Markets, American Family Investments; and Jane Carpenter, Associate, Quantitative Research Group within Ares Global Client Solutions contributed to this piece.