By Dr. Jim Dahle, WCI Founder
I often see bizarre beliefs and statements regarding investment returns. If you want to be an intelligent, functional lifelong investor, you must understand how investment returns work.
In this post, we'll go over many of the truths about investment returns that you need to know so you can mentally combat the falsehoods constantly being told about them.
#1 Most People Have No Idea How to Calculate an Investment Return
Go ahead. Ask the next 10 people you see to calculate an investment return if you provide them with the necessary data to do it. How many pulled it off? None did. Few people know how to calculate a return. Several different returns exist, but the one most commonly used to compare one investment to another is the Internal Rate of Return or IRR. Perhaps the easiest way to do this is to take all of the cash flows into and out of the investment and input them into a spreadsheet like Microsoft Excel or Google Sheets and use the XIRR function. It looks like this:
Contributions are positive. Withdrawals are negative. You have to use the “DATE” function in column 2. And the last row is the current value of the account (as a negative number) and today's date. The return it spits out is “annualized,” so if your time period is less than a year, it'll be higher than your actual return.
Since most people have no idea how to calculate their returns, don't believe them when they tell you their returns. I don't even believe those managing money for me when they tell me MY returns until I calculate them myself. You might be surprised by how many discrepancies I see.
#2 Beware the Cocktail Party Phenomenon
When you go to a cocktail party, all sorts of topics come up. However, no matter whether people are talking about dating, travel, or money, they will generally only talk about their wins. The losses are just too painful to share, and we fear they'll make us look stupid. Nobody wants to look stupid. In fact, we don't even tell ourselves about our own losses. We tend to repress them from our memory until they are forgotten. Don't forget about this when hearing others talk about their returns.
More information here:
The Nuts and Bolts of Investing
9 Ways to Increase Your Investing Return
#3 The Only Returns That Matter Are After-Tax, After-Fee, and After-Inflation
An investor has three very serious enemies: taxes, fees, and inflation. Most returns are reported as pre-tax returns. This is understandable since we all have different tax situations. Some returns are even reported pre-fee for some bizarre reason, and almost nobody adjusts returns for inflation. Yet the only returns you can spend are after-tax, after-fee, and after-inflation. Get used to adjusting any returns you see for these three factors.
#4 Many Investors Ignore the Value of Their Time
Investors often engage in an activity that is best thought of as a combination of work and investing. The investor is inputting their time and effort as well as their money. Classic examples are day trading and direct real estate investing. If there is any significant input of your time or effort, you should adjust your returns for that. It just isn't fair to say, “I forced appreciation of 40% of my investment property last year,” without mentioning the six weekends you spent knocking out a wall, building a new bedroom, and renovating a bathroom.
When calculating, you can use the value of your time doing something else (practicing medicine?) or the value of what it would have taken to hire someone else to do that work. But don't pretend all that money was just due to you making a wise investment with your nest egg.
#5 Return and Risk Are Related
When you hear people talk about incredible returns, don't forget about the risk they were taking to get those. Investments that go up 50% in a year tend to also go down 80% in a year. Now, taking high risk is by no means a guarantee of high returns—even in the long run—and plenty of risk is completely uncompensated. But you're unlikely to get high returns without taking significant risk. The risk-free investment is basically what you can earn on cash in a high-yield savings account or money market fund. As I write this, that is about 4% nominal. With inflation running right now (as measured by CPI-U) at just under 3%, that risk-free return is in the 1%-1.5% real (after-inflation) range. If you need to make more than that to meet your goals, you're going to have to risk the potential loss of principal, both in the short term and the long term.
Stocks have long-term returns of about 10%, but they go down in value about 1 out of 3 years, with drawdowns of 50%+ about once a generation and a 90% drop in the Great Depression. If that's what it takes to get 10%, imagine what kind of loss you're risking to get 15%, 20%, or 30%?
More information here:
Risk vs. Reward — How to Find the Balance
#6 Trees Don't Grow to the Sky
People love to extrapolate past returns into the future. However, just as trees don't grow to the sky, some returns simply cannot continue at the same level as the value of the business or cryptoasset or whatever increases.
Let's use Bitcoin as an example. At the beginning of 2011, you could buy a Bitcoin for 30 cents. At the beginning of 2025, that same Bitcoin cost you $94,420. That's an annualized return of 147%. If Bitcoin continued to appreciate at 147% for six more years, it would be worth more than everything else in the world. Suffice to say that Bitcoin will not continue to appreciate at that rate because it cannot continue to appreciate at that rate. Trees don't grow to the sky.
Same thing with businesses. WCI's profit 5Xed in one year, and the company value increased 5X. Between the income and the appreciation, that was a 600% return that year. It has never happened again. Even the most successful companies in the history of the world don't grow at that rate. Apple had a very good 2024. It grew earnings by 12%. Not 600%.
#7 Performance Chasing Is Not Only Common, But Typical
If you think you're not susceptible to performance chasing, you may be right. You probably aren't right, though, since most people are VERY susceptible to it. We buy stuff that has recently gone up in value, hoping it will go up more. Sometimes it does, sometimes it doesn't. But in general, the technique leads to buying high and selling low, classically ineffective investing behavior.
I can't believe how many experienced investors I saw justifying an increase in their allocation to US large growth tech stocks in early 2025. They claimed all kinds of reasons for doing so besides the obvious. You know, things like a 53% return in 2023 followed by a 29% return in 2024, giving a 5-15 year annualized return of 19%-21%.
Design a reasonable, diversified plan and stick with it. Don't change it due to a nasty bear market, and don't change it due to one of your asset classes having an incredible run for a few years. Both are likely to lead to regret eventually.
#8 Very High Returns Do Not Persist and, In Fact, Cannot Persist
Being a student of financial history is a worthwhile activity. It doesn't take much studying to get a sense for what reasonable returns are. The US has been the most successful country with long-term stock market returns. From 1871-2024, the historical return on US stocks has averaged 11%. That annualizes to 9.36%, including dividends. If you adjust for inflation, that drops to 7.09%. So, a reasonable return to expect for taking significant risk is 9%-10% a year nominal, 7% real (after-inflation). Add some leverage risk, and maybe you can boost that up into the 10%-15% range, a place where real estate investments can live.
Yet I keep seeing people in the physician real estate space claiming they can do returns of 20%, 30%, 40%, or more long-term. That's not true. In fact, it can't be true.
Let's do a little exercise. Let's start with $1 million and have it grow at 40% a year every year. How long until you own the entire world? The sum of global wealth is $454 trillion. If you start with $1 million and get 40% a year, you hit $454 trillion in just
=NPER(40%,0,-1000000,454000000000000) = 59 years
That's right, a single person's investment horizon. Do you know anyone who owns the entire world? I mean, the richest person in the world as I write this is Elon Musk with $482 billion. You'd pass him at year 38. And this is without any additional contributions to the account. No, if someone tells you they can get you a 40% return long term, they're lying to you. Take Warren Buffett, one of the most successful long-term investors in the history of the world. What is his annualized return from 1965 to the present? Just under 20%. And over the last 30 years? Just 10%. That's what success looks like. Not 40%. If you can reliably beat the stock market by just 1%-2% long term, you should be managing billions, not just your own money. If you can beat it by 30%, you should soon own the world.
The next time you see a slide with someone talking about how they're going to get 25% or 30% or 40% returns, you can just roll your eyes and move on to the next presentation at the conference.
More information here:
Why You Can’t Invest by Looking in the Rear-View Mirror
10 Ways to Console Yourself When Losing Money in the Markets
#9 You Don't Become Successful by Chasing High Returns
I've met a lot of very successful people in the decade and a half I've been doing The White Coat Investor thing. Almost universally, they have become so rich in the same very boring way.
- They make a lot of money.
- They carve out a big chunk and put it toward wealth building.
- They invest in some reasonable fashion and stick with it.
I did run into one guy online who claimed to have made $50 million in crypto by the time he was in his mid-20s. But that is not the way to bet on getting wealthy. That's almost as bad as buying lottery tickets. Yes, it's possible but not likely to get wealthy that way, and if your habits don't change, it's unlikely you'll stay that way for long. Stop trying to shortcut the process.
Yes, if you're the entrepreneurial sort and you work hard, take some risks, have a high savings rate, and add a little leverage, maybe you can become financially independent in 5-10 years instead of 10-15. I am convinced that the fastest reasonably reproducible route out of medicine is a small empire of short-term rental properties you manage yourself—at least for a little while.
Yes, many docs still haven't acquired much wealth after 20, 30, or even 40 years of practice. But the difference between those docs and the successful ones is not generally their investments. Lots of people wonder whether to pay off debt or invest or how much leverage to keep in their lives. In my experience, the successful people do both. The same people who save a lot of money are also the people who pay off their debt. The people who don't pay off debt are also the ones who don't save much and don't have much to invest.
Even if you get 40% returns for a year or two, it isn't going to matter if there isn't much invested. I frequently see people with six-figure net worths lecturing people with eight-figure net worths about how to invest using other people's money. Maybe if you had some money of your own, you wouldn't have to use other people's money.
The Bottom Line
You need to understand how returns work, how to calculate them, and what returns to expect so you aren't misled by your peers, financial professionals, and scammers.
What do you think? What do you think people don't understand about returns? What else should people know about their returns?
I like the examples of Buffet, Bitcoin, and WCI: the biggest returns show up in the first 10-20 years. However, no one can predict which investments will be the next big thing. It’s basically like angel investing, but you have to realize that some (most?) of those investments will lose money. Only a few will really take off. Will that make up for your other losses? Maybe, maybe not.
One of the things I’ve been wondering about lately is with the large influence of private equity now, most companies come public after all the big gains have been made, if they even go public at all (e.g. OpenAI, which is private and it’s current valuation is one of the 20 largest companies). At this late stage, only then do investors in the public markets have an opportunity to invest in these high growth companies, but only after the founders and VC types have made their outside gains and the level of growth available to the public markets is really diminished.
In the past, raising capital in the public markets was one of the main methods for funding growth in relatively large businesses. Now it seems there is so much cash sloshing around, that there are plenty of ways for companies to fund their growth and they don’t really need the public markets to raise money any more. In addition, there is the added hassle and reporting requirements that private companies don’t have to be concerned about. I wonder what kind of equity returns can be expected in the future as compared to what we have been accustomed to the last several decades. I fear our children will not have the same opportunity to have the same equity returns we have come to expect. Of course the future is unknown, but I really wonder about the current state of affairs and the likely future returns.
Crystal ball cloudy. And the private/public pendulum swings back and forth over years. Doesn’t seem worth lying awake at night worrying about this to me given how little control you have over future returns.
I really like #1, #3, and #4.
#1 reminds me of when I recently learned about the different ways interest is calculated (e.g., simple, compound, CAGR, IRR, and APY). It’s fascinating how the method you use can completely change your understanding of growth over time.
#3 highlights how many investors overlook their fund’s expense ratios or underestimate how much a 1% annual fee can cost in dollars, especially over a 10-year or longer period. It also brings to mind the importance of tax-efficient asset location and the impact of inflation on purchasing power.
#4 reminds me of the book Your Money or Your Life by Vicki Robin and Joe Dominguez, which drives home the idea that managing money well isn’t just about returns, it’s about aligning your financial choices with your values and time.
In short, these truths are timeless. Thanks for posting them!
This is why I simplify our plan as much as possible for spouse’s sake. Aside from his early “put our money in that stuff that went up 20% in a year” (me incorrectly getting him into a mutual fund one good market year before he wanted it all out to buy a new car- ie he shouldn’t even have been in a stock fund! The same error with my best friend had her lose 10% a different year and stop taking my investment advice!!!) and his more recent argument that a 5% CD rate for only a 6 month CD (teaser rates from the credit union, rest of the rates are more realistic) should be listed as a 2.5% rate and 6% for 8 month should be advertised as 4%. Sometimes I wonder if he’s pulling my leg… like when I warned him of the ‘windfall’ he’d get from term life if I died in childbirth and he said “Gee I could buy a plane. Or invest in race horses.”
Dr. Dahle, you have taken all the rest of us to task for not knowing how to calculate IRR. One constructive thing you could do is to define what you mean by IRR, which you did not do in this article, but which is not hard to do. A second thing to note is that if you yourself have to rely on a black box spreadsheet function that someone else has written to get the answer, you do not really understand it either. Unless you can write the function yourself. One unsettling fact is that in some circumstances, one can show that the calculation can yield multiple legitimate answers for the IRR, r1, r2, …. I wonder what the spreadsheet does in this circumstance, pick one of the returns r1, r2, etc. at random and return it without comment? Tell you the calculation failed?
I”m not sure that # 1 my math skills are strong enough to write the function myself nor that # 2 most in the audience are interested. Sounds like a good subject for a guest post:
https://www.whitecoatinvestor.com/contact/guest-post-policy/
This gets into the mathematical weeds in ways most WCI readers probably wouldn’t enjoy, but it’s true that the IRR equation can have more than one mathematically correct answer and Excel isn’t “solving” it exactly. The issue is that we’re looking for the discount rate r that makes the sum of a polynomial of degree N (the number of periods) equal to zero, and such an equation can have up to N solutions.
If there’s just one negative cash flow (the initial investment) followed by positive ones (dividends, interest, or eventual sale proceeds), then there’s a single real solution for r. But most investors are adding to investments repeatedly (eg, monthly purchases of VTSAX in a 401(k)) before eventually taking withdrawals, which complicates the cash-flow pattern.
My understanding is that Excel’s IRR and XIRR functions use the Newton–Raphson method (Google it), which is a numerical technique. It starts with a guess, uses the function value and its derivative to improve the estimate, and stops once the results converge.
I still agree with Jim that using Excel’s XIRR is far better than what usually passes for people’s “investment return” claims — but it’s worth remembering that it’s an approximation, not an exact algebraic solution.
Which perhaps explains why one sometimes gets an “error” on it despite entering info correctly.
Excel gives an error message if it can’t find a solution after 20 iterations of Newton-Raphson. You can use MIRR (Modified Internal Rate of Return) instead.
I’ll have to try that one. I don’t get an error often, but it does happen every now and then.
We agree that it is important to know your IRR (internal rate of return). But no one has defined it, or told the reader exactly what it is. If you ask google,
it returns word salad.
Here is my attempt: The IRR is meant to apply to an investment where the return may not be constant in time, like a stock fund, where today’s return is generally different than yesterday’s. The investor may add or remove money from the account at various times. Suppose the investor had a hypothetical account with a constant (fixed) return r, continuously compounded, and made the same additions and subtractions as the actual account. If r is adjusted so that the final value of the hypothetical account is equal to the final value of the actual account, then r is the IRR.
Example 1: If the only contribution or withdrawal from the account is an initial contribution x0, and the final value is xf, over a time period T, then the IRR is given analytically by r = (1/T) ln ( xf / x0 ), where ln is the natural log function. (The value of the hypothetical account increases exponentially over time as x = x0 exp( r * t ), where exp is the exponential function.)
Example 2: If the only contributions or withdrawals from the account are made at the beginning and at the halfway point T/2, there is an analytic solution for the IRR r given by the quadratic formula from high school algebra. The quadratic formula gives two solutions. It may happen that one of the
solutions is a complex number, not a real number, and so can be rejected as unphysical.
Example 3: If the only contributions or withdrawals are made at the beginning and the T/3 and 2T/3 points, (at the one third and two thirds points), there is an analytic solution for the 3 solutions for IRR. Similarly, if the only contributions or withdrawals are made at the beginning and the T/4, the T/2, and 3T/4 points, there is an analytic solution for the 4 solutions for IRR. Again, it may happen that some can be rejected because they are complex, rather than real numbers. In the general case, there is no analytic solution, but a solution can be found numerically and iteratively by Newton’s method (or another method). There is no reason to distrust the numerical solution. It will be accurate to many more decimal places than you care to look at. But it is possible that there will be more than one real solution for r. I do not know how to choose among them.
Please correct me if something I wrote is wrong. My field is physics, not finance.
Discrete compounding is used much more often than continuous compounding for IRR so the underlying equations don’t use the natural log:
0 = Σ_{t=0}^{n} CF_t / (1 + r)^t
I think you are concentrating too much on the math rather than the purpose of using IRR over other calculations for the return. This is an exercise in descriptive mathematics, not precision engineering. The point of using the IRR is that we can estimate a return after a variety of flows both in and out of an investment and account for the time-based value of money as opposed to just saying, “my retirement account is worth twice what I contributed.” It is not to find the node points where an analytic solution is possible. I care about this because the IRR of my equity portfolio is around 11% as opposed to 3% for a bond portfolio, not because I can prove my IRR is exactly 11.45% which beats someone else’s of 11.27%.
There is no practical difference between compounding daily and continuously. Use whatever you prefer.
The main issue for me was that no one in this thread (and elsewhere) had so far defined IRR clearly enough that one could calculate it within a factor of two or ten, except by using black box software that runs on an algorithm that no one could describe. If this topic does not interest you, skip ahead.
If your investment grows at a rate r, saying that it grows exponentially and giving the very compact formula means something to some people. Saying that it grows in some jerky stair-step fashion that cannot be described without a full pad of paper or a computer means very little.
I loved this article, as it shines a light on the myriad shortcuts and mental gymnastics we often use to justify our investments. We all need to take a sober and accurate look at our returns (and how they were made) from time to time. I think points 1, 2 (although I’ve always called this the Doctor’s Lounge Phenomenon), 5, and 6 are especially important.
The only one I partially disagree with is #4. While I completely understand and appreciate Dr. Dahle’s perspective, I think a little more nuance is needed. As I started my career in Emergency Medicine, I also began to invest in real estate directly. I spent a lot of time hanging drywall, laying tile, and painting, creating the foundation for what is today, eight figures of rental real estate equity.
At that time, I was also working more than 20 shifts a month. All my old scrubs have paint on them because I so often went straight from the hospital to a rehab site. My point of contention is with how Dr. Dahle values this time. Comparing it to my hourly rate in the ER seems to be a false equivalence, since mentally, I couldn’t have safely worked any more shifts. I also didn’t have a wife or kids at the time, so the true equivalence would be sitting on my couch watching TV, which doesn’t pay very well. However, since I could listen to real estate, personal finance, and investing podcasts while laying flooring or painting, I received my 10,000 hours of education in these topics while performing tasks I considered relaxing compared to the stresses of the ER.
As mentioned in the article, I could calculate the comparable cost of someone else doing the work, but then I would never have learned skills that still benefit me to this day. I think of this time more as an education, like attending trade school, than work I could just sub out.
The benefit of a hands-on approach isn’t limited to real estate—the lessons learned from starting and running a business compound with time. Even if the first attempt isn’t successful, you carry the experience with you for the next attempt. I’m sure Dr. Dahle spent countless hours learning things he applied to growing WCI. Some of this work is obvious and can perhaps be calculated in a return, but much of it was probably more intangible and harder to quantify; yet still valuable.
I know I’ve set a lot of conditions for Dr. Dahle to be wrong and me right: #1 you must max out the hours at your job (which is way more than most people think), #2 you must enjoy your extra activity and find it relaxing, #3 the “juiced” returns must be worth the squeeze compared to your net worth, and #4 you can’t have a higher/better use of your time (like a family). These conditions may only be met for a season of your life; I stopped regularly doing these activities once that season changed for me, so I think it’s important to take advantage of the time you have.
While I don’t think anyone should overestimate their investment returns by discounting the time spent building sweat equity, I also don’t want to discourage young physicians from pursuing these activities. While I agree that you shouldn’t “pretend all that money was just due to you making a wise investment with your nest egg,” I believe that the education and experience you gain from sweat equity can pay dividends far beyond what can be easily calculated in an IRR (if you can even figure out how to calculate one correctly).
High returns don’t persist too long, but you don’t need them to if you have enough invested to make it count to you. If it happens that a company matures and has lower returns but is still rock solid business-wise (competitive advantage maintained, etc), you are lucky indeed. The people who need to keep chasing high returns are those who sell their winners when they get to 5% or 10% of their portfolio. You know, mowing the flowers and watering the weeds. The whole point of investing IMO is to not need high returns for too long.
I find it fascinating you want to have this discussion here, instead of on this page:
https://www.whitecoatinvestor.com/how-to-calculate-your-return-the-excel-xirr-function/
Maybe you didn’t know that other page existed.
Not sure what you mean. As a guess you think I should want to quibble over the meaning of the word ‘high’.
The comments system sometimes doesn’t link comments up the way they’re intended once there start to be multiple comments and replies. It’s just not as good as what you might see in true forum software or something like Reddit.
This particular response was directed at trugs and his extensive discussion of the math behind IRR. Not sure why it ended up where it was and appeared to be directed at you.
I did not know about that page. But as I look it over, it also does not describe IRR clearly enough that one could write an algorithm to calculate it. This is a major omission. As far as I know, I am the only one on any WCI thread who has attempted to do so.
And why would someone attach a comment to a column that is over a decade old, rather than to a current discussion on the same topic?
Yes, I’m pretty sure you’re the only one. And based on participation, certainly the person who cares the most about looking behind the curtain.
Consider the question, are there any actual examples where the calculation of the internal rate of return r (IRR) can go wrong? One enters the dates and amounts of deposits and withdrawals to an account, and the final value of the account. Using the definition above, one would like the algorithm to return a single real value for r, complex numbers are not allowed. Can this fail? Can there be no solutions, or more than one solution?
If you are not interested in this question, you can stop reading now. (This may mean you PharmMedMD, and perhaps others.)
The IRR r can be a negative number, for example if you make an investment and it loses money. If there are only contributions, no withdrawals, and one enters a final value for the account that is a negative number, there is no (real) solution for r. If there are only contributions, no withdrawals, and the final value of the account is a positive number, one can show that there is a unique (real) solution for r. (If a nonzero amount of time elapses between the last contribution and the time at which the final value is taken.) This is good, the case we want. [From the fact that if a function is strictly increasing, it cannot go through zero more than once.] A silly case: if all contributions and withdrawals are zero, and the final value is zero, there are an infinite number of solutions, any (real) r is a solution. Under the same conditions if the final value is nonzero, there are no solutions.
Are there any nontrivial examples where there is more than one real solution, r1, r2, perhaps more? Yes. Consider the case where a deposit x1 is made at time t=0, a second deposit x2 is made at time T, and x3 is the final value at time 2T. This case can be solved by the quadratic formula from high school algebra. In all cases for nonzero x1, there are two solutions, r1 and r2. However, one or both solutions may be complex, so they don’t count. When are there two real solutions? One can show that this occurs when
-4 * x1 * x3 < x2 * x2, and
case A: x1 is positive, x2 and x3 are negative, or
case B: x1 is negative, x2 and x3 are positive.
To get two real solutions for r, one of the contributions must be negative. In fact, case B is like the careers of many doctors. One initially borrows money to go to school (negative initial contribution), later invests a positive amount, and eventually gets back to zero and then a positive net worth.
Case B example, x1 = -1, x2 = 3, x3 = 1 (in units of anything, so x2 = 3 could be $3,000, or $300,000). If T = 10 years, the solutions for r are r1 = .09624 and r2 = -.09624 (negative), roughly plus or minus 9.6%. It seems that r1 makes more physical sense. Where are you going to find someone who will lend you money at a negative interest rate? (Although interest rates in major international bond markets have been negative recently.) This suggests that one could get rid of some troublesome multiple solutions by being suspicious of someone lending you money at a significant negative interest rate.
If someone could tell me, if one plugs these numbers into their favorite IRR algorithm, does it return just the positive number r1 and ignore r2?