Compounding Hates Friction

Imagine you have a penny. Every day, it doubles. Day one: ¢1. Day two: ¢2. Day ten: $5.12. By day thirty, that coin is worth over $5.37 M. You know the math. You've seen the slide. Compounding is the eighth wonder of the world — or so the quote goes.

But here is what that slide never shows you. What happens if, each time your coin doubles, you hand 10% of it to the government first?

Day one: ¢1 becomes ¢1.90 instead of ¢2. A small difference. Negligible, even. But by day thirty, your 5.37M penny is only worth $1.21 M!! You didn't lose 10% of your wealth. You lost 77% of it. And you paid the same tax rate the whole time.

That is not a tax story. That is a geometric failure story.

Most investors think about tax drag the wrong way. They see a 15% capital gains rate and think: I lose 15 cents on every dollar of gain. That framing is almost completely wrong. The real cost of tax is not the rate. It is when the rate is applied — because every dollar sent to the government today is a dollar that stops compounding on your behalf forever.

Charlie Munger explained this more cleanly than most economists ever have. If you buy something that compounds at 15% for thirty years, and you pay a single 35% tax at the very end, your after-tax return is 13.3%. Not 15%. Not 9.75%. 13.3%. The deferral itself is worth 3.5% per annum. Every. Single. Year. For thirty years.

Now run the alternative. Same investment. Same 15% growth. But you pay 35% tax annually on your gains. Your after-tax return collapses to 9.75%. A $1,000 investment grows to $43,000 in scenario one. In scenario two? $16,000. The investment is identical. The business didn't change. The market didn't change. Only the timing of when the government collected changed.

Malone did not become one of the most effective wealth compounders in American history by finding better assets. He found better structures. He understood that the government is, as he puts it, your partner from birth — they just don't get to come to all the meetings. The goal is to delay those meetings for as long as legally possible.

Tax drag is not arithmetic erosion. It is exponential erosion. The standard future value formula — FV = PV(1 + r)^n — assumes the full return stays invested. The moment you introduce an annual leakage, you are not just reducing the growth rate. You are shrinking the base that next year's growth compounds on. And the year after. And the year after that.

The table below makes the mechanism concrete. Start with $1,000 growing at 10% annually for thirty years.

Look at the bottom row carefully. The deferred investor and the annually-taxed investor at 35% started with the same dollar, in the same asset, earning the same return. After thirty years, one has $17,449. The other has $4,383. That is a 4:1 ratio in terminal wealth from nothing more than the timing of tax collection.

For a high-net-worth investor with a $10 million portfolio, a 2% annual tax drag does not cost them $200,000. It costs them roughly $4.5 million in lifetime wealth — permanently removed from the compounding cycle. Not temporarily. Not recoverable. Gone.

Here is a question most investors never think to ask: at what point does trading most damage your returns? The intuitive answer is: the more you trade, the more you pay. So the damage is proportional. Roughly linear.

The research says otherwise. Robert Arnott's work on after-tax returns revealed something startling: the first 10% of annual turnover accounts for more than half of the total value destroyed compared to a zero-turnover strategy. The damage front-loads. The marginal cost of the first trade is enormous. The marginal cost of the tenth trade is much smaller.

To understand that, we need to go back to what turnover actually does in a taxable account. Every time you sell an appreciated position, you convert a deferred gain into a recognized one. You hand money to the government that was previously compounding on your behalf. The new position you buy then has to overcome a performance hurdle just to get back to where you were. Arnott estimates that hurdle is approximately 4.1% to 6.6% of extra annual return — depending on the holding period — just to break even after tax.

Most active managers never clear that hurdle. Not because they are bad at picking stocks. Because the tax friction of activity is structural, and the return premium needed to overcome it is consistently underestimated.

Warren Buffett does not view deferred tax liabilities as a burden. He views them as an interest-free loan from the U.S. Treasury, repayable at his election. That framing is not creative accounting. It is precise.

In his 1989 letter to Berkshire shareholders, Buffett illustrated the point simply. If $1 doubles every year for twenty years:

Under annual taxation at 34%: your $1 grows to $25,250.

Under a single tax at the end at 34%: your $1 grows to $692,000.

Same tax rate. Same investment. 27:1 difference in terminal wealth.

Charlie Munger extended this logic to Wesco Financial's balance sheet, where the company carried a significant deferred tax liability. Munger argued — correctly — that the DTL was not a liability in the conventional sense. It was float: capital reinvested productively for decades, with the tax bill deferred indefinitely. He estimated the liability was worth roughly $30 per share of intrinsic value, not zero.

The most tax-efficient businesses in the world are those with reinvestment moats: high ROIC paired with long runways to deploy capital internally. Companies like Constellation Software or Dino Polska that reinvest nearly 100% of operating cash flows into organic growth at high returns are not just good businesses. They are tax-deferral machines. Their intrinsic value compounds internally, invisible to the tax collector, year after year.

Charlie Munger was fond of inversion: don't ask how to succeed; ask how to fail, then avoid it. Applied to compounding, the question is not how do I maximize my return? It is how do I avoid destroying it?

The answer is almost always the same. Do not trade for the sake of doing something. Activity is a cost, not a demonstration of diligence. Do not prioritize reported earnings over owner earnings and free cash flow — the map is not the territory. Do not ignore the performance hurdle: if you sell today, the new asset must earn roughly 4–6% more per year just to get you back to where a held position would have been. And do not mistake urgency for insight. Market orders in volatile periods pay the spread. Patience earns it.

The tax code, Munger noted, gives a patient investor an extra 1% to 3% per annum — for free — simply by holding. In the world of decades, that is not a rounding error. That is the difference between wealth and a legacy.

The real question is not whether you understand compounding. Most investors do. The question is whether you have internalized that compounding is a geometric process, and friction is geometric too — working against you at the same rate your returns work for you. The coin doubles every day. The question is how much of it you keep.