Distributions to Paid-In (DPI)

Definition

DPI (Distributions to Paid-In) measures the total value that has been returned to investors as actual cash (or cash equivalents), relative to the total capital they paid into the fund.

DPI = Cumulative Distributions / Paid-In Capital

Key thresholds:

  • DPI < 1.0× – The fund has not yet returned investors’ original capital in cash
  • DPI = 1.0× – Investors have received back exactly what they put in; any further distributions are profit
  • DPI > 1.0× – Investors have received their capital back plus cash profit above it

DPI is sometimes called the “money-in-pocket multiple” or “realisation multiple.”


Why It Matters

DPI is the most reliable performance metric for alternative investment funds because it is based on actual cash received, not estimates. Unlike TVPI (which includes the GP’s estimate of remaining portfolio value), DPI cannot be inflated by optimistic valuations.

For LPs, DPI matters practically:

  • It represents real liquidity – cash that can be redeployed or used to meet obligations
  • It removes valuation uncertainty: a distribution has actually occurred, at a known value
  • It is the primary basis for assessing whether a fund has “returned capital” to investors, which governs waterfall mechanics and carried interest eligibility

DPI is often described as the hardest metric to manipulate. A GP can adjust portfolio valuations; they cannot easily invent distributions that did not occur.


What “Distributions” Includes

DPI counts all forms of value returned to LPs, including:

  • Cash distributions – Standard wire transfer of proceeds from an exit
  • In-kind distributions – Shares in a listed company (e.g., after an IPO) distributed directly to LPs rather than sold by the fund. Once distributed, the shares are at the LP’s risk.

Note on in-kind distributions: When the fund distributes listed shares, the LP bears any subsequent share price movement. If the shares fall in value before the LP can sell, the effective DPI is lower than when the in-kind distribution was booked. Some LP systems record in-kind distributions at the distribution date price; others track the eventual sale price.


Worked Example

A PE fund has called £80m (out of a £100m commitment) and has made two distributions:

EventAmountCumulative Distributions
Exit of Company A (Year 4)£25m£25m
Partial exit of Company B (Year 6)£25m£50m

DPI = £50m / £80m = 0.625×

Investors have received 62.5p for every £1 paid in. The fund holds remaining positions valued at £90m (RVPI = £90m / £80m = 1.125×; TVPI = 1.75×).

The fund looks good on TVPI, but only 62.5p of actual cash has come back per pound invested. Whether the RVPI is achievable will determine whether TVPI or DPI is the more accurate indicator of final performance.


DPI Through the Fund Lifecycle

DPI is typically very low in the early years (the J-curve period) because exits take time to materialise. It rises progressively as investments are realised.

Red flag signals:

  • Low DPI in a fund over 8 years old suggests exits are slow or delayed, which may indicate portfolio problems or adverse market conditions
  • Rapid early DPI growth (Year 1–3) may indicate quick recapitalisations rather than true exits – investigate the nature of the distributions

Key Limitations

  • DPI does not include unrealised value. A fund can have excellent TVPI but modest DPI if exits have not yet occurred. Neither figure alone tells the complete story.
  • DPI does not reflect time. 1.0× DPI in Year 3 is exceptional; 1.0× DPI in Year 10 means the fund returned capital but no profit (in cash).
  • In-kind distributions add complexity. The value of an in-kind distribution depends on when and at what price the LP sells the received shares – which is outside the GP’s control.
  • Recycling provisions can suppress DPI. If the fund recycles proceeds back into new investments rather than distributing them, DPI stays lower for longer even if exits are occurring. See recycling provisions.


Manual Review Notes

  • The assertion that DPI is “hardest to manipulate” is conceptually sound (actual cash vs estimates) but GPs can time distributions strategically to influence period-end metrics. This nuance could be added in a future revision.
  • In-kind distribution accounting varies across fund administrators and LP systems. This section gives a simplified treatment; operational teams should verify how their specific systems handle this.