What it is
GARP stands for growth at a reasonable price. It's the discipline of buying companies that are genuinely growing and genuinely well-run — but refusing to pay any price to own them. GARP investors want the compounding that comes from a strong, expanding business, and they want it at a valuation that still leaves room to make money.
It's a deliberate middle path. Deep value buys cheap and tolerates mediocre businesses; pure growth buys fast growers and tolerates nosebleed valuations. GARP takes the best of each and rejects the worst: a great business bought expensively can be a bad investment, and a cheap business going nowhere can be a value trap. GARP insists on both a quality, growing company and a sensible price.
GARP = a good, growing business × a fair price
The label is most associated with Peter Lynch, who ran the Magellan fund to a legendary record by looking for growing companies whose valuation hadn't yet caught up to their prospects — often summed up by the PEG ratio (see Understanding the PEG ratio).
Why it works
GARP works because it collects two independent edges at once and cancels the weakness of each:
- Quality and growth compound. Businesses with high returns on capital that reinvest into more high-return growth get more valuable over time, almost mechanically. Owning them lets time do the heavy lifting.
- Valuation discipline caps the downside. The fastest way to lose money in good companies is to buy them at a peak multiple. Paying a reasonable price protects you when growth slows or sentiment turns — the multiple has less room to fall.
Deep value can languish for years when the market rewards growth; pure growth gets crushed when expensive multiples compress. GARP is less exposed to either extreme, which historically has meant steadier, more repeatable returns rather than the highest possible ones in any single regime.
What "good" looks like
A textbook GARP name usually shows:
- Durable growth — mid-teens or better revenue and earnings growth that looks sustainable, not a one-off (see Understanding revenue growth).
- Real quality underneath — high returns on capital, healthy margins, and strong free cash flow, not growth funded by debt or dilution (see Understanding quality investing).
- A reasonable multiple — a PEG near or below 1, or valuation multiples that are modest relative to the company's growth and its sector, not the whole market.
- A sound balance sheet so a bad year doesn't threaten the business.
The word doing the work is reasonable. GARP does not require a bargain — it requires that the price be fair given how good and how fast-growing the business is.
Common gotchas
- Paying up for a "quality" story. The most common failure is quietly dropping the price discipline and rationalising a rich multiple because the business is wonderful. That's just growth investing wearing a GARP costume.
- Cheap-but-broken. The opposite failure: chasing a low multiple on a company whose growth is fake or fading. Reasonable price on a deteriorating business is a value trap.
- Cyclical mirages. A cyclical company can look like cheap growth at the top of its cycle, right before earnings roll over. Normalize across the cycle.
- Growth that isn't paid for in cash. Revenue growth funded by stock issuance or debt isn't the same as self-funding compounding. Check that cash flow keeps pace.
How we use it
GARP is the Bull Rankings model. Every US stock we screen gets one transparent quality-growth score from 0–100, built on three pillars that are exactly the GARP recipe:
- Quality — profitability, returns on capital, and balance-sheet strength.
- Growth — revenue and earnings growth, and whether it's durable.
- Value — the valuation you're paying, graded against sector peers so a fast grower isn't punished on absolute multiples.
A high score requires all three: a strong, growing business at a fair price. In honest out-of-sample testing, dropping the value pillar and chasing quality-growth alone actually reduced returns — direct evidence that the "reasonable price" half of GARP is doing real work, not just sounding prudent. Names that are wonderful but expensive aren't errors; they're routed to the separate Compounders shelf, where the price discipline is relaxed on purpose. You can see the highest-scoring GARP names right now on the screener or by sector (for example, technology or healthcare).
Bottom line
GARP is growth at a reasonable price — owning good, growing businesses without overpaying. It works because quality and growth compound while valuation discipline limits the damage when things wobble. Demand all three — quality, growth, and a fair price — mind the trap of quietly paying up, and let the compounding do the rest. It's the whole idea behind the one score on this site.
GARP is the whole model here, not one screen among several
This site runs a single score, and it is a GARP score. That's a design choice with consequences worth stating plainly.
The score combines three pillars — quality, growth and value — using a floored geometric mean rather than a weighted average. That distinction is the mathematical expression of GARP.
An average lets a name compensate: score 95 on growth, 20 on value, and the average still looks respectable. A geometric mean does not. A near-zero on any pillar drags the whole result down no matter how strong the other two are. A company has to be acceptable on all three to rank well, which is precisely what "growth at a reasonable price" means — you are not buying the fastest grower, and you are not buying the cheapest stock. You are buying the intersection.
Where the "reasonable price" is judged
Within the value pillar, the growth-adjusted reads lead deliberately:
| Read | Weight |
|---|---|
| PEG vs sector | 1.0 |
| P/E vs sector (only if PEG unavailable) | 1.0 |
| P/E vs the stock's own history | 0.7 |
| FCF yield vs sector | 0.4 |
| EV/EBIT vs sector | light tilt |
PEG leads because a raw multiple can't distinguish 30× for a business growing 30% from 30× for one growing 5%. The non-growth-adjusted reads are kept as tilts specifically so they cannot override the case the screen exists to find: a quality grower that looks expensive on a flat multiple and reasonable once growth is priced in.
Every one of those reads is a percentile against sector, not an absolute threshold. There is no "PEG under 1" rule, because what growth is worth differs legitimately across industries and a fixed cutoff quietly turns a stock screen into a sector bet.
Guards stay light on purpose
A GARP screen can be ruined by protecting it too well. Every filter that keeps out a bad company also keeps out a good one, and the failure mode of an over-guarded screen is that it stops finding anything interesting.
So the model's guards are deliberately loose. The accruals read — a known value-trap signal — is a light tilt at weight 0.4 rather than a gate, because working-capital builds naturally lift accruals at exactly the fast-growing companies a GARP screen should be surfacing. Concentration limits cap how many names come from one sub-industry rather than excluding sectors. The DCF isn't a guard at all: it doesn't enter the score, and sits beside it on the stock page as a cross-check.
The principle is that the pillars should do the work and the guards should catch only what is obviously broken.
What GARP will and won't get you
It will systematically miss the extremes. The cheapest stocks in the market are usually cheap for a reason the quality pillar will penalize. The fastest growers are usually priced accordingly and the value pillar will mark them down. If your thesis is deep value or hypergrowth, this model is the wrong instrument and will argue with you constantly.
It is not a market-timing tool. The score says a business is sound and reasonably priced. It says nothing about when the market will agree, and the backtest's record — ahead of its benchmark in 7 of 13 calendar years, with most of the lead coming from three of them — is the honest evidence of that.