Business Growth

The Growth-Investment Decision: Evidence on When to Reinvest vs Distribute

Every profitable quarter ends with the same unasked question: where should this money go? Most founders answer by default - reinvesting out of ambition or distributing out of fatigue - rather than by framework. The evidence says the stakes are high. McKinsey's long-run analysis found companies in the top third of resource reallocation delivered a 10.8% annual return to shareholders versus 2.5% for the least active reallocators, compounding to a six-fold market-value difference over two decades (McKinsey, 2014-2017 research line). Bain's research finds only about one company in nine sustains profitable growth across a decade, with 85% of the barriers internal rather than market-driven (Bain, 2016). This article translates the capital-allocation evidence - including Warren Buffett's one-dollar retention test - into a working decision system for service-business owners.

Joshua Agonya Pi'Rwot

By Joshua Agonya Pi'Rwot

Founder, Business Growth Accelerator

Executive summary

Only about one company in nine sustains profitable growth for a decade, and reallocation discipline separates them from the rest. This research review gives founders a working framework for the reinvest-versus-distribute decision.

Section 1

The five challenges at a glance

Founder capital allocation fails in five characteristic ways. Profits are reinvested by inertia into last year's activities regardless of returns; growth bets are made without any hurdle rate, so capital flows to enthusiasm; the cash buffer is treated as dead money and left chronically thin; distributions are either reflexive - starving compounding opportunities - or absent, leaving owners over-concentrated in a single illiquid asset; and no one revisits past allocations, so losing bets run indefinitely. The table summarizes the evidence base. Sources here are comparatively strong: McKinsey and Bain publish long-horizon strategy research, the Federal Reserve and JPMorgan Chase Institute provide independent small-business data, and Berkshire Hathaway's shareholder letters are primary-source doctrine on owner capital allocation. Reading down the root-cause column reveals the common thread: every failure is a missing rule, not a missing insight. Founders generally know their legacy line is tired and their pet project is struggling; what they lack is a pre-committed mechanism that converts knowledge into reallocation before another year of inertia compounds. That is why the framework in this article emphasizes written policies - buffer targets, hurdle memos, experiment caps, and distribution formulas - over case-by-case judgment, which the evidence shows defaults to repetition (McKinsey).

Section 2

Challenge one: inertia is the default allocator

The most consistent finding in the capital-allocation literature is that organizations do not actually allocate - they repeat. McKinsey's multi-decade analysis of large companies found a third reallocate a mere 1% or so of capital year to year, with the average at 8%, while the top third of active reallocators earned 10.8% compound annual total returns to shareholders versus 2.5% for the most static - a gap that compounds to roughly six times the market value over 23 years (McKinsey, 2014-2017 research line). The mechanism scales down to a 12-person agency intact: this year's budget starts from last year's, every service line and role defends its incumbent funding, and the implicit assertion - that last year's allocation remains optimal - goes untested. For founders the inertia is emotionally reinforced: the legacy service line is identity, the long-tenured role is loyalty, the familiar marketing channel is comfort. Bain's complementary finding sharpens the cost: 85% of the barriers preventing sustained profitable growth are internal and controllable, not imposed by markets (Bain, 2016) - inertia chief among them. The corrective is procedural, not motivational: once a year, build the allocation from zero. List every place a discretionary dollar could go - service lines, hires, marketing channels, tooling, buffer, distribution - and force each to bid for capital with an expected return, including the incumbents. The point is not spreadsheet precision; it is breaking the presumption of incumbency.

Section 3

Challenge two: reinvestment without a hurdle rate

Reinvestment has a halo - it sounds like ambition and feels like virtue - but the evidence treats it as a claim requiring proof. Warren Buffett stated the test in his 1984 shareholder letter: unrestricted earnings should be retained only when there is a reasonable prospect, backed preferably by historical evidence or thoughtful analysis, that every dollar retained will create at least a dollar of market value for owners (Berkshire Hathaway, 1984). For a private service firm, the translation is direct: a retained dollar must generate more owner value inside the business - through durable profit growth or enterprise-value appreciation - than it would in the owner's hands after distribution. That bar is higher than it sounds. Bain's research shows only about one in nine companies sustains a decade of profitable growth while earning its cost of capital (Bain, 2016), which means most incremental growth spending, across the economy, fails the dollar test. The practical discipline is a founder-scale hurdle rate. Before any material reinvestment - a new hire ahead of demand, a service-line launch, a rebrand - write down the expected payback period and the evidence supporting it, then compare against the honest alternative: what the same dollars earn distributed and invested in diversified assets. Reinvestments with proven unit economics - sales capacity into a converting pipeline, delivery automation with measured time savings - clear the test routinely. Reinvestments justified mainly by narrative momentum usually do not, and naming that distinction is the entire discipline.

Section 4

Challenge three: the buffer and the owner are claims, too

Founder allocation frameworks habitually treat the operating business as the only destination for profit, with cash reserves dismissed as lazy money and distributions as a guilty pleasure. The small-business evidence inverts both judgments. JPMorgan Chase Institute's transaction-level research found the median small business holds just 27 cash buffer days - 31 in professional services (JPMorgan Chase Institute, 2016) - and the Federal Reserve's 2026 employer-firm survey shows the environment that buffer must absorb: rising costs as the dominant financial challenge, revenue declines slightly outnumbering gains for a second consecutive year, and roughly a third of financing applicants facing funding gaps (Federal Reserve Banks, 2026). In that world, the buffer is not idle capital; it is purchased optionality - the ability to hold pricing in a soft quarter, retain key staff through a client loss, and buy distressed opportunities competitors cannot. Its return is invisible until the day it is the only thing that matters. Distributions earn their place by a different logic: concentration risk. A founder with every dollar of net worth inside one illiquid, key-person-dependent firm is running a portfolio no adviser would design. Systematic distributions that build outside assets reduce personal risk and - counterintuitively - improve in-business decision quality, because a founder with a funded personal balance sheet can decline bad revenue, fire misfit clients, and play long games. The allocation stack, in order: fund the buffer to target, then reinvest what clears the hurdle, then distribute the remainder without apology.

Section 5

Innovative solutions

Several practices from institutional capital allocation adapt cleanly to founder scale. The annual reallocation summit: one day per year, the founder - ideally with a fractional CFO or peer board - rebuilds the capital plan from zero, with every incumbent spend re-bidding alongside new proposals, directly attacking the 1%-reallocation inertia McKinsey documents (McKinsey). Stage-gated growth bets: new service lines and market expansions are funded like a venture portfolio - small proof-of-concept tranches with pre-written success metrics, follow-on capital released only when gates are met, and a pre-committed kill decision if they are not. This imports McKinsey's finding that systematic, rules-based reallocation outperforms episodic conviction (McKinsey, 2017). The written dollar test: every reinvestment above a set threshold gets a one-page memo - expected return, payback period, evidence, and the distribution alternative - creating the historical record Buffett's test asks for (Berkshire Hathaway, 1984). Buffer laddering: reserves are tiered - operating cash, a 60-to-90-day core buffer benchmarked against the 27-day median (JPMorgan Chase Institute, 2016), and an opportunity reserve explicitly earmarked for downturn acquisitions and talent grabs. And the owner's quarterly dividend policy: a fixed distribution formula - for example, a percentage of trailing profit above buffer target - which removes both guilt and impulse from the decision and makes reinvestment the deliberate exception rather than the unexamined default.

Section 6

Solution framework: the four-bucket allocation system

Run every dollar of profit through four buckets in strict order, quarterly. Bucket one, resilience: fund the cash buffer to target - 60 to 90 days of operating expenses for most payroll-heavy service firms, set against the documented 27-day median (JPMorgan Chase Institute, 2016) and the cost pressures in current Fed data (Federal Reserve Banks, 2026). Until this bucket is full, nothing else gets funded. Bucket two, proven reinvestment: opportunities with demonstrated unit economics - capacity behind a converting pipeline, automation with measured payback, pricing and packaging work with tested demand. These clear Buffett's dollar test on evidence, not hope (Berkshire Hathaway, 1984). Bucket three, experimental bets: capped at a fixed percentage of profit - many operators use 5-15% - deployed in stage-gated tranches with pre-written kill criteria. This is where new service lines audition without endangering the firm; Bain's one-in-nine base rate for sustained profitable growth is the humility built into the cap (Bain, 2016). Bucket four, distribution: everything remaining flows to owners by formula, building the outside balance sheet that de-risks the founder and disciplines the firm. Twice a year, audit the system itself: did bucket-two investments hit projected paybacks; did bucket-three gates actually trigger kills; did the buffer get raided for non-emergencies? The audit is what separates an allocation system from an allocation story - and what the evidence says most firms never do (McKinsey; Bain, 2016).

Section 7

Evidence-based action plan

Month one: establish the baseline. Compute buffer days, list every discretionary spend above your materiality threshold, and tag each as resilience, proven reinvestment, experiment, or legacy inertia. Most founders find the largest category is the last one - capital allocated by habit, which McKinsey's research identifies as the silent performance killer (McKinsey). Month two: install the rules. Set the buffer target (60-90 days), the experimental cap (5-15% of profit), the reinvestment hurdle with written one-page memos, and the distribution formula. Write them down; unwritten policies dissolve under the next exciting opportunity. Month three: run the first reallocation summit. Rebuild the budget from zero, force incumbent spending to re-bid, and make at least one genuine reallocation - the research is unambiguous that movement, not analysis, is what separates the 10.8% compounders from the 2.5% laggards (McKinsey). Quarterly thereafter: run profits through the four buckets in order and review experiment gates. Annually: audit realized returns against the memos, Buffett-style - did each retained dollar create a dollar of value (Berkshire Hathaway, 1984)? Expect the honest answer to retire some convictions; that is the system working. The strategic endgame is a firm that compounds deliberately - resilient enough to survive shocks, disciplined enough to fund only real returns, and generous enough to its owner that nobody needs the business to be everything. For adjacent evidence in this pillar, see [Profitable Growth Has Replaced Growth at All Costs: The Evidence and the Operator Playbook](/blog/growth-profitable-growth-replaces-growth-at-all-costs) and [Bootstrap or Raise in 2026: The Research on Outcomes, Control, and the Selective-Capital Middle Path](/blog/growth-bootstrap-vs-raise-2026-selective-capital).

FAQ

Direct answers for operators.

How much profit should a service business reinvest versus distribute?

No universal ratio survives contact with the evidence; the right split falls out of a sequenced system. Fund a 60-90 day cash buffer first - the documented median is just 27 days (JPMorgan Chase Institute, 2016) - then reinvest only in opportunities with evidence they return more than a dollar per dollar retained (Berkshire Hathaway, 1984), cap experiments at 5-15% of profit, and distribute the rest by formula.

What is Buffett's one-dollar test for retained earnings?

In his 1984 shareholder letter, Warren Buffett wrote that unrestricted earnings should be retained only when there is a reasonable prospect - backed preferably by historical evidence or thoughtful analysis - that every dollar retained will create at least a dollar of market value for owners (Berkshire Hathaway, 1984). For private firms, the parallel is whether retained profit grows enterprise value and durable earnings more than the owner could earn investing a distribution elsewhere.

Does the research really show reallocation matters that much?

Yes, and over long horizons. McKinsey's analysis spanning 1990 to 2013 found the most active third of capital reallocators compounded total shareholder returns at 10.8% annually versus 2.5% for the least active - roughly a six-fold market-value difference over the period - while a third of companies moved only about 1% of capital each year (McKinsey). The mechanism, budget inertia, operates identically in small firms.

When should a founder kill a growth investment?

At the gate you wrote before funding it. Stage-gated bets with pre-committed success metrics and kill criteria outperform conviction-based persistence, because Bain's research shows only about one in nine companies sustains profitable growth and 85% of failure causes are internal and manageable (Bain, 2016). If a bet misses its gate, the pre-written decision executes - renegotiating with yourself mid-failure is how pet projects are born.

Joshua Agonya Pi'Rwot

Written by

Joshua Agonya Pi'Rwot

Founder, Business Growth Accelerator · Country Director, AVODA Group Uganda · EMBA

Joshua helps service-business operators turn scattered marketing into a clear path from first attention to booked call. He is Founder of Business Growth Accelerator and Country Director of AVODA Group Uganda.