Cash, leverage, debt, equity, and the difference between fake and real opportunity costs

Worth doing with cash

Many years ago when I spent more time on the travel hacking circuit, I heard a guest talking about a then-current Office Depot/OfficeMax deal. She exclaimed, “it was worth doing with cash!” This made such an impression on me that it’s stuck with me through the years, and has exerted a kind of gravitational pull of commonsense that not all travel hackers are naturally endowed with, myself very much included.

Longtime readers will recall that over the course of the merger between Office Depot and OfficeMax, the stores continued to be operated independently (Madison, Wisconsin, where I lived at the time of this anecdote, had an Office Depot on one side of town and an OfficeMax on the other), but they had synchronized their weekly coupons. A common promotion at the time offered a $15 discount off the purchase of $300 in Visa or MasterCard gift cards (depending on the week).

The promotion worked at both stores, but the stores had different inventory and sales policies: OfficeMax had fixed-value gift cards that could be purchased with credit cards (like Chase Ink cards that earned 5 Ultimate Rewards points per dollar) and Office Depot had variable-value cards that could only be purchased with cash and debit cards. What this meant was that most people were stuck buying two $200 fixed-value cards with their Chase Ink cards, paying two activation fees, and earning a small profit before credit card rewards and liquidation costs.

The speaker at this gathering was making the simple point that $500 variable-value cards were worth buying without earning any credit card rewards at all: you could earn all your profit up front by buying $500 variable-value cards with cash, and not face any purchase restrictions at all; she could clear the store out, write them a check, and walk out with $1,000 in profit (after liquidation). As I recall, in this iteration of the promotion purchasing three $500 gift cards would trigger five $15 credits, for a pre-liquidation profit of $54.15 per 3-card purchase.

I still use this anecdote as a kind of quality control check when I’m exploring a new technique, or a new approach to an old technique, because it helps me break them into individual parts: which parts cost money? Which parts earn money? Is there a way to skip the parts that cost money and jump straight to the parts that earn money?

A simple example is bank account signup bonuses. Doctor of Credit curates an endless list of these bonuses, which typically let you earn a few hundred dollars for setting up a direct deposit and sometimes leaving some money in the account for a few months. Some of these banks also allow you to fund your initial bank account deposit with a credit card, usually limited to a few hundred or thousand dollars.

You can imagine someone taking these two facts and deciding “the best technique is to use my highest-earning credit card to fund accounts with banks with the highest new account bonuses that accepts credit card funding.”

In this example the error is clear: you should sign up for the highest new account bonuses regardless of whether they can be funded with credit cards. Accepting a $200 signup bonus instead of a $300 signup bonus so you can charge $500 to a credit card that only earns 2% in rewards is the purest false economy — use cash and earn the $300 instead.

Leverage

If ordinary Americans ever encounter the idea of leverage, it’s in one of two places: the purchase of a primary residence, and the gutting of the industrial and commercial giants of the 20th century. The principle is the same in both cases.

Given any amount of starting capital (including $0), you can earn an excess return on any investment through the addition of borrowed money: $100,000 spent on a $100,000 house allows you to earn the appreciation on a $100,000 house, but $100,000 spent on the down payment of a $500,000 house allows you to earn the appreciation on a $500,000 house. If you can’t make your mortgage payments, or the house craters in value, you can walk away with nothing but a stern warning on your credit report.

Likewise, in the leveraged buyout of a stodgy old industrial firm, investor groups can borrow against the assets of a company and strip it for parts. If they are able to unload the shell to some greater fool at a profit, then they’ve earned free money on the exercise. If not, they haven’t staked any of their own money on it to begin with anyway.

This is the core of the ideology of leverage: stake as little of your own fortune as possible to secure an equity interest with unlimited upside.

Against opportunity cost

Many pillars of elementary economics, which is as far as most business and finance journalists seem to make it, rely on the conceit of rational economic actors who are constantly operating at an “opportunity frontier.” That is to say, all their resources are optimally allocated across all available opportunities at all times. When a new opportunity comes along, the economic actor determines whether and how many resources to allocate to it by comparing it to all other existing opportunities.

I like to think of this as the “bump off” fallacy: each new opportunity is evaluated to see whether it should bump a lower-earning opportunity off the bottom of the stack of available opportunities.

The problem, which I have only become more aware of as I have become more experienced, is that virtually no one operates anywhere close to the opportunity frontier. This is frequently parsed by malicious people as “this generation would rather eat avocado toast than save for a down payment,” but I am making a much more literal argument: most people do not hold even their long-term cash savings in accounts that earn the maximum rate of interest available to them.

They are not guilty of anything for not maximizing their rate of return. There are only so many hours in the day, and too many people spend too many of them between working and getting back and forth from work. You would be insane to blame them for preferring dinner, a beer, and a football game than moving money back and forth between high-interest checking accounts.

The flip side of that is that “opportunity cost” is an absurd way to describe the choices of someone who is not operating anywhere close to the opportunity frontier.

You only get to keep the difference

There is a commonsense logic behind the concept of opportunity cost, however, which I use when making decisions about using cash or debt for any given opportunity: you only get to keep the difference.

That is to say, if an opportunity comes along that lets you earn (pulling a number out of thin air) 19.6% APY (1.5% per month annualized), and you choose to finance that opportunity with cash that’s currently earning 5.01% APY, then you shouldn’t take credit for the full 19.6% you see in returns, but only the difference between that and the return you would have earned on your status quo ante, a “mere” 14.59%.

Obviously I’m being a bit facetious: that would be an excellent thing to do with your cash savings, if you had to. But if you didn’t have to, and you could keep the whole whack, you’d be even better off.

Administered interest rates on credit card debt make a lot of things worth doing with debt

Back in November I described my distinction between “administered” and “market” interest rates. Credit cards offer a curious combination of the two: an administered interest rate of 0% on loans of between 15 and 50 days (depending on when the purchase falls in the statement cycle and the length of your grace period), and a (high) market interest rate on loans after each statement’s grace period has expired. A credit card earning 2% in cashback rewards offers even more favorable terms, the equivalent of a 2% yield, rather than interest charge, on each new loan you originate each month, as long as you pay it back in time.

Let me illustrate this with an example present at this very moment in the real world. The Giant grocery store chain is currently offering 2 or 3 points per dollar on the purchase of Vanilla Visa prepaid debit cards. Their variable-value cards still cost $6.95 (while Safeway’s cards have gone up to $7.95 in my region), so each purchase yields a profit of $3.05 or $8.05 in grocery rewards ($10 or $15 in grocery rewards less a $6.95 activation fee), minus any liquidation cost. This is a classic “worth doing in cash” situation: the profit is present however you pay for the gift card that generates it.

But remember, you only get to keep the difference. If you did pay with a debit card linked to an account earning 5% APY and it took 5 days to liquidate the gift card back into your balance, how much did it cost you? About $0.33 in interest, bringing your final profit down by the same amount.

In this situation, the rationale for using a credit card is obvious regardless of any rewards earned: why sacrifice the interest earned on your cash when you have access to a costless source of credit?

Inverting the logic of leverage

The common examples of leverage I offered, residential home purchases and corporate buyouts, have in common the basic conceit that acquiring ownership of a productive asset with borrowed money lets you secure a larger future income stream than you would be able to afford with your cash on hand, while enjoying an unlimited upside and capped downside.

What I am telling you is that these opportunities do not only exist at the tail end of the duration curve, the 30-year mortgage or the 10-year airline turnaround or the 5-year electric car plant retooling. The precise same logic applies at very shortest durations, where administered interest rates make debt free or profitable to take on.

When and where that is the case, using capital rather than debt to pursue the same opportunities puts a drag, rather than a tailwind, on your total investment return.

Upton Sinclair, petty capitalism, and intrusive thoughts

“It is difficult to get a man to understand something, when his salary depends upon his not understanding it.”

I’ve been thinking about that line from Upton Sinclair’s campaign memoir “I, Candidate for Governor,” on and off for the last year, and I’ve come to believe it gets one thing very right and one thing very wrong.

Cultures are very good at reproducing themselves

It’s currently fashionable to claim that Americans are in a maelstrom of loneliness, but what I’m continually struck by is the astonishing plethora of civic organizations you can see on every streetcorner. You may suspect that the Elks, Eagles, Pachyderms, and Masons are aging and insignificant because you happen to not enjoy spending time with them, but whenever I visit small and medium-sized towns they’re the first places I see teeming with activity. I’ve gone to bingo at the VFW, sung the Montana state anthem at a Kiwanis meeting (“Montana, I love you!”), and danced lacklusterally in ballrooms at the Elks Club.

If you don’t like those examples, let me suggest you look at the stately manors lining the most valuable real estate in the most successful cities in America: note they all seem to have Greek letters outside, because they’re owned and occupied by the members of fraternal organizations that have managed to survive not by stringing along aging members into senescence or renting out their ceremonial halls for funerals, but by recruiting brand new troops of 18-year-olds each and every academic year.

These cultures adapt and shift over places and times, but the reason they perpetuate isn’t because of their adaptability, it’s because of the urgency of their internal logic. And, in fact, most human activities take this form.

The culture of petty capitalists is determined by the nature of petty capitalisms

If you have ever known someone before and after they acquire a rental property, you’re no doubt familiar with the sudden change in personality they undergo. Suddenly obsessed with trash pickup schedules, noise ordinances, and lawn maintenance, the simple act of holding a deed transforms your laid-back friend into the most annoying landlord you’ve ever met, overnight.

The same pattern repeats everywhere. Anyone who has bet money on the Superbowl can transform before your eyes into a specialist in NFL rules mechanics. Any piker who busts a hand in blackjack will suddenly be able to explain why the player to their right screwed them by making the wrong play and, really, that two of hearts should have been theirs, if you think about it.

I recently read an anecdote [I thought this anecdote was from Gay Talese’s memoir “Bartleby and Me” but I can’t find it there, so consider this hearsay instead] of a door-to-door furniture salesman from the 40’s or 50’s who explained to the reporter that the only thing worse than a customer who couldn’t make their installment payments was one who paid off their furniture early. After all, a late payment just added to their total interest, most of which went to the salesman as commission, but an early repayment meant cutting short the entire future stream of income from the exhausting work of closing a door-to-door furniture sale on installment.

What all these examples have in common is that they take an outcome you might otherwise be neutral towards (landlords versus tenants, Texas versus New England, furniture debtors versus furniture creditors) and, purely through the mechanism of self-interest, make you emotionally and morally invested in one side over the other.

This is a mistake, but it’s a very easy mistake to make, so one shouldn’t feel too bad about it, but rather strive to do better.

Prosper and the strange case of “prepayment risk”

If you’ve ever heard of “prepayment risk,” it was probably in the context of mortgage loans. Lenders financing long-term, fixed-rate loans live in the same terror as our door-to-door furniture salesman: if a loan can be repaid at any time, then it’s impossible to predict the loan’s future stream of income. If short-term rates rise, then the lender is stuck collecting below-market interest on their capital. But if rates fall, then the borrower can refinance the loan at a lower rate and force the lender to reinvest their capital at a similarly lower rate.

Since ordinary people borrow money to buy houses, rather than lend money to others to buy houses, most people understand that this is a feature, rather than a bug, of our housing finance system. Since financiers cannot live in houses, but people can, people should be free to refinance their loans as often as it’s advantageous while rates are falling in order to more easily afford shelter, but not be forced to refinance their loans when rates rise, since that would make their shelter less, rather than more, affordable.

What I was not expecting from investing in peer-to-peer loans with Prosper is the way it so closely mirrors the experience of a mortgage lender and the similarly intrusive thoughts it generates.

From a lender’s perspective, prepayment risk is the dominant risk on the Prosper platform. This should surprise you. After all, Prosper makes extremely risky loans. Here’s one offering a yield of 29.25% on a 5-year loan:

View fullsize

This loan will probably not be repaid in full and on time, which Prosper itself explicitly acknowledges (the historical return on such notes is shown as 2.62% - 13.18%). If you had some way of only picking the 29.25% loans that would be repaid in full and on time over 5 years, you’d make a killing only investing in those notes. But you don’t, so some of the loans you pick will default and the more loans you buy, the closer your actual returns will get to the overall return of similarly-risky notes.

Prepayment risk is the flip side of default risk, because Prosper also makes extremely safe loans, and people with good credit and steady income have much cheaper alternatives for credit than unsecured Prosper loans. In the same way that people who just-barely qualify for Prosper loans can quickly slip into default, people who easily qualify for Prosper loans can quickly find more favorable terms from other lenders.

Since I started investing with Prosper again in January, 2023, 37 of my loans have been paid in full, compared to 488 in current repayment, 5 with late payments, and one charge-off. In other words, while investing in highly-rated loans, prepayments are a vastly more frequent occurence than settlements, bankruptcies, and charge-offs.

Why do Prosper lenders hate prepayments so much?

There’s something intrinsically funny trying to explain why Prosper lenders hate prepayments. Here’s a typical example of a prepayment. The loan was originally for 4 years at a 14% yield:

View fullsize

The borrower made their first two payments on-time, then repaid the remaining interest and principal balance:

View fullsize

I made a total of $0.72 ($0.77 in interest less $0.05 in Prosper’s “service fees”) on a $25 loan, a mere 2.88% total return, far from my promised 14% yield. But I made it in 2.5 months (October 23 to January 7), not 4 years, which through a little exponential arithmetic gives me an annualized return of 14.6% (this is not to be taken too literally; on small amounts over short periods each penny and each day shows up as a big difference in annualized yields).

There are essentially three reasons that I think Prosper lenders spend so much time fuming over prepayments. First is a sense of powerlessness. Prosper determines the eligibility of borrowers and the repayment term and interest rate they’ll pay. Lenders have to accept or reject the loan terms agreed on by Prosper and the borrower. Since the lender is subject to the whims of the Prosper lending algorithm, lenders have a sense that borrowers should be too. If a lender is willing to surrender access to their money for a full 5 years, why shouldn’t the borrower be locked into the same repayment period?

Second is a sense of scarcity. Once Prosper has determined the loans that are available, it feels to lenders like they’re doing valuable work sorting through those loans to determine the most qualified borrowers and most favorable repayment terms. Prosper invites this mania by listing obviously meaningless information like the borrower’s state of residence and information about their residency, credit history, income, and employment. The provision of this information mechanically instills in the lender the conceit that they can apply some novel combination of metrics to select the loans most likely to be repaid on time and in full, but not a moment sooner. If you’re carefully hand-picking loans and one of them gets repaid after just a few weeks or months, then you’ve not only lost that dedicated stream of interest revenue but you’ve also wasted all the time you spend hand-picking that loan.

And finally, there is a perfectly accurate sense that Prosper gets paid first, because they do: their loan origination fee (“1%-9.99%, depending on Prosper Rating”) is subtracted from loan proceeds before they’re distributed to borrowers. This means Prosper gets paid generously to process the churn from quickly-repaid, high-quality loans: the same $25 loan that netted me the $0.72 in interest over 2.5 months generated roughly $2.55 in revenue for Prosper (a 1% origination fee and $0.05 in servicing fees).

Of course, Prosper incurred expenses managing the loan, and I didn’t.

You cannot keep the intrusive thoughts out but you don’t have to agree with them

Hopefully I’ve given enough examples in this post that you recognize yourself or someone you know in at least one of them. The fact is, people succumb to moralizing when they are personally financially involved in a way they never would about identical financial transactions they don’t have a stake in. Once you become a realtor you are compelled to believe that a 3% buyer’s fee was handed down to Moses on Mount Sinai. Once you become a New York City apartment broker (a profession that everyone agrees does not and should not exist anywhere else) you are instantly convinced that one month’s rent is the correct and natural broker’s fee.

What I am here to tell you is that you cannot resist thinking those thoughts, but you do not have to agree with them. We all know that there are good people, who resist the intrusive thoughts of their profession and adhere to a moral code independent of their income source. We all know there are normal people who are just trying to get by in whatever profession they happened to land in. And we all know there are bad people, who succumb instantly to the imperatives of their profession.

So, when you feel those imperatives, do you resist them, go along with them, or succumb to them? Do you think homeowners have special insight into the unfairness of property taxes just because you happen to be a homeowner? Do you think employers have special insight into the unfairness of the minimum wage just because you happen to be an employer forced to pay the minimum wage? You can’t keep from thinking such thoughts, but you are not forced to agree with them when they intrude on you unbidden.

Two years investing with Prosper peer-to-peer lending

For the last few years I’ve been scaling up my investments with the Prosper peer-to-peer lending platform. Since there is essentially no community information available (the Prosper forum on reddit has been silent for 7 years, and Bogleheads was even less help), I was accumulating datapoints from scratch. Now that I’ve spent some dedicated time on the platform, I want to share how investing with Prosper really works.

My investment parameters

As I mentioned back in May, I have Prosper’s “recurring order” function (not to be confused with “Auto Invest”) set up to invest in $25 increments in “AA”- and “A”-rated notes with yields of 10% and higher. This has two valuable functions. First, when I deposit cash in the account it is automatically invested as qualifying notes are added to the platform. Second, since interest and principal payments are added to the same cash balance, it keeps repayments from sitting idle and therefore contributes to compounding discipline. Whenever deposits or repayments result in $25 or more in my cash balance, it’s invested in the next qualifying note that’s added to the platform.

The ineffability of rates of return

At this point we have to make a digression into the metaphysical. Annual percentage yield (APY) is a common and useful metric of investment returns because, when properly calculated, it offers a like-for-like comparison between investment products.

If you read that sentence again, you’ll see “when properly calculated” is doing all the work, and in point of fact, APY is really only useful in a few specific contexts, like bank accounts and long-term zero-coupon bonds, where your entire balance earns the specified interest rate and interest payments are automatically reinvested at the same underlying rate. In other contexts you may calculate a sort of pseudo-APY to compare products, but it should not be confused with your actual, real-world rate of return.

This is obvious to anyone who thinks about it for a moment, so there are lots of alternate methods of calculating return which are useful in their individual contexts. The “SEC return,” which you often see in descriptions of mutual funds, is a set of formulae the SEC has authorized to be reported based on dividend payments. Likewise “dividend yield,” usually seen in stock and ETF listings, is a mechanical calculation based on the company’s most recent dividend payments and current share price.

Somewhat more exotically, you have the question of actively-traded funds and hedge funds which hold some of their investors’ money in cash. How should the return of these funds be calculated? Based on the total amount investors have committed to the fund, or on the actual investments the fund makes? Should investors count the cash held in a hedge fund waiting to be invested against their cash allocation or their investment allocation? Should their rate of return be calculated only on invested cash, or on the total balance held with the fund?

These questions aren’t mathematically hard, they’re hard because there’s no right answer: you should use the calculation that gives you the most useful answer for the question you’re asking.

This brings us to Prosper’s return calculation.

The Prosper return calculation

Before we continue, let me quote Prosper’s description of their return calculation at length. This return appears both on the app and desktop website (emphasis mine):

“Your Return is calculated using a formula where (A) the numerator is equal to the sum of all the interest received on active Notes, plus all late fees received on active Notes, minus all servicing fees paid, minus all collection fees paid, plus all net recoveries received on charged-off or defaulted loans, plus all net debt sale proceeds received on sold loans, plus all net sale proceeds received on Notes sold on Folio, minus gross principal losses, plus any investor promotions credited to your account; and (B) the denominator is equal to the sum of the amount of active principal balance outstanding at the end of each month since account opening. The results are then multiplied by 12 to get an annualized return. This gives us the ‘Return’ for your Prosper Notes.”

Within this paragraph we get a few answers to how Prosper answers the metaphysical questions I posed above. For example, we’re told that uninvested cash does not count towards the return calculation: only your invested cash (“active principal”) does. But the part I highlighted describes a different confusion you will encounter as you scale up your Prosper investments: invested principal counts towards your return calculation before a single payment is due.

Remember, Prosper makes individual loans to borrowers, and you are investing in a slice of those repayment-dependent notes. After a loan has become funded by investors, the money is distributed to the borrower. The borrower’s repayments start at least a month after you’ve put in your order for a slice of their loan. But until the first payment is made, your slice of the note counts only towards the denominator of Prosper’s return calculation.

This means Prosper’s return calculation will always be depressed by the amount of active principal not in repayment in any given month. This is a feature when it reflects a loan with late payments or that has been sent into settlement or bankruptcy, but it is misleading when it reflects an eventually-performing loan that simply has not begun repayment yet.

Your Prosper return gets more accurate as loans age into repayment

To illustrate this effect, I completely stopped making new deposits into my Prosper account in October, so that all my existing notes would have a chance to enter repayment (or become past due). Only money reinvested from repayments of interest and principal would affect the denominator of my Prosper return, and sure enough, you can see the return wobble depending not on the performance of my loans, but on the amount of my total balance that reflects new loans. This chart shows my Prosper return over 2024:

View fullsize

Those returns are recorded as:

• January: 10.58%

• February: 10.52%

• March: 10.53%

• April: 10.02%

• May: 9.05%

• June: 8.23%

• July: 10.19%

• August: 8.18%

• September: 9.19%

• October: 9.53%

• November: 9.04%

• December: 10.16%

The essential thing to understand is that this wild range, from 8.18% to 10.58%, does not reflect any underlying economic reality. It only reflects the amount of new principal invested each month, which contributes to the denominator of the return calculation, but is not reflected in the numerator until months later.

Calculate your own return based on mature notes

Fortunately, Prosper provides detailed account statements, which are usually available in the mobile app around the 5th or 6th of each month, and on the desktop website a bit later. These statements let you calculate your own rate of return depending on the metrics that matter to you. My preferred method is to base returns on the previous month’s “Principal Balance of Active Notes.” All of those notes should have entered repayment by the end of the following month, so it lets me focus exclusively on the interest earned on mature notes, disregarding uninvested cash and notes too new to have begun repayment. Here are my 2024 returns based on that metric (each month is shown as an annualized rate of return):

• January: 17.2%

• February: 12.8%

• March: 13.6%

• April: 15.5%

• May: 13.8%

• June: 12.4%

• July: 13.6%

• August: 11.3%

• September: 12.8%

• October: 12.3%

• November: 11.2%

• December: 12.8%

As you’d expect, the biggest differences between the return Prosper calculates and the return I calculate take place in the months where I’m purchasing notes most actively (July-September), and the smallest differences occur in the months I let my notes age into repayment (November and December), but my Prosper return will always be below my own calculation because Prosper will only show your return based on the entire lifetime of your account, and I have had my account for many, many years.

The significance of the early years, when I liquidated my notes on their resale platform at a loss (I was broke and needed the money), will be diluted over time but will never disappear from my Prosper return calculation unless and until they update their formula.

A note on “tax efficiency”

I am now required by the conventions of financial journalism to say something about tax efficiency, the idea that you shouldn’t care about the actual returns you earn from your investments, but rather how much of those returns go to your local, state, and federal governments.

I consider this a deeply unhealthy attitude that causes people to do preposterous things with their money, but the wisdom of our ancestors is in the simile; and my unhallowed hands shall not disturb it, or the country’s done for.

Interest payments on notes held in Prosper’s standard taxable account are taxed at your marginal income tax rate. Return of principal (78% of my December payments, for example) is untaxed, and losses of principal (bankruptcies, settlements, and charge-offs) are deducted from your interest payments to arrive at your total taxable interest amount.

One approach to tax efficiency might be a tax-exempt municipal bond mutual fund. Glancing at Vanguard’s list of state-based tax-exempt mutual funds, current SEC yields range from 3.14% for California (VTEC) and 3.83% for Pennsylvania (VPALX). Those yields are much lower than my returns to date on Prosper, but you can imagine someone investing on Prosper with a 50% marginal tax rate and 50% default rate ending up more or less breaking even between the two. My marginal tax rate is much less than 50% and my default rate is much less than 50% so I do not consider this to be a realistic concern. A high-income person investing in the riskiest loans is free to reach their own conclusions.

An approach I have more sympathy for is making Prosper loans in an Individual Retirement Account, which Prosper is set up to faciliate. Only cash can be contributed to IRA’s, so you can’t move your existing notes into an IRA, but you can contribute up to your annual limit and I assume you can move cash in from existing IRA’s, although I’ve never tried.

Traditional IRA’s allow you to deduct your contribution from your taxable income each year, but much more importantly allow your investments to compound tax-free inside the account. I do not know of a single person in history who has literally sold off their investments to pay their tax bill, but if you’re that person, then having your Prosper investments compound tax-free will save you the trouble of finding investments to liquidate to pay the taxes on your interest.

Another reason to consider this is that most investments people hold in their tax-advantaged retirement accounts are already quite tax-advantaged. Vanguard’s LifeStrategy Growth Fund (VASGX) only issued 2.3915 cents per share (0.054% of Net Asset Value) in short-term capital gains in 2024; the rest was taxable at the heavily-discounted long-term capital gains rate. Holding those long-term assets in a tax-advantaged account is a waste of the internal tax-free rate of return, which is best spent on highly tax-inefficient investments like Prosper loans.

The flip side is the difficulty of liquidating Prosper loans, inside or outside of tax-advantaged accounts. Once you have reached your risk tolerance for such loans, then you’d naturally want to turn off the reinvestment features and start moving the resulting interest payments into other investment vehicles. But tax-advantaged accounts, unlike taxable accounts, make this process onerous and increase the drag on returns of your uninvested cash.

To summarize, an IRA is the ideal place to hold Prosper loans because of their inherent tax inefficiency and the benefit of internal tax-free compounding, but actually doing so requires a level of sophistication and diligence that the overwhelming majority of people should not be expected to demonstrate.

Conclusion

I have much more to say about Prosper, which I surely will at a later date, but that seems like enough to chew on for now.

As I mentioned in my earlier post, the three most important risks to consider when investing with Prosper are platform risk (Prosper declares bankruptcy and your notes become unsecured claims against Prosper, not the individual borrowers), underwriting risk (Prosper got something fundamentally wrong when underwriting notes and they are in fact much riskier than they are purported to be), and macroeconomic risk (the notes are underwritten properly but an economic crisis drives even well-qualified borrowers into default).

Those risks are real. If they were not real Prosper would not offer returns 3-5 times higher than the risk-free rate on US Treasuries. But Prosper does offer such returns, and if you’re willing to bear those risks, you’re able to earn outsized (pre-tax) returns, until one or more of those risks comes to fruition.

If that happens soon, you’ll lose some or all of your principal. If that happens a long time from now, or never, you’ll make a killing. And that, to me, is the essence of investing.

Robinhood never understood what their appeal was

[edit 11/25/24: a trusted reader in the comments has the 3% cash back credit card, so it does really exist]

I’ve written a number of times about Robinhood, the free stock-and-ETF trading app, over the years, and back in March I wrote about what I described as “one of the best all-in-one financial products out there.” I was wrong, so today I want to explain both how I was wrong and why it matters.

Recap: the Robinhood Gold pitch

When I wrote that post, I described the new $5-per-month Robinhood Gold service as consisting of two unrelated components:

• a 5% APY interest rate on uninvested cash

• and a 3% cash back credit card (this product, to the best of my knowledge, still has not been launched, so set it aside for now).

Some people pointed out that this product also includes $1,000 in free margin in your investment account, but as will become clear, that is not worth nearly as much as it sounds. I would go so far as to say it is worth nothing, or less than nothing if it causes you to enroll in Robinhood Gold.

Market rates versus administered rates

My first mistake was misunderstanding how Robinhood markets the interest rates on uninvested cash, which illustrates the difference between “market” rates and what I call “administered” rates.

A “market” rate is like the one set on your credit cards or adjustable-rate mortgages: your loan agreement specifies a particular market index reported in some reputable newspaper and says you’ll pay that rate plus-or-minus an adjustment based on your down payment, credit rating, or the phases of the moon. The point being, the rate will change but it will change in a specified way known in advance: when the published market rate rises or falls.

An administered rate is set by the policy of the institution offering the investment or holding your funds. It is also subject to change, but it is set to administrative change, not mechanical or contractual change.

For example, US government Series EE savings bonds have an administered interest rate of (at least) 3.53% APY: the bond is guaranteed to pay out twice the invested amount after 20 years. If interest rates rise above 3.53% APY while you’re holding the bond, you also have the option to cash it out (with a penalty) and invest in the higher-yielding asset instead.

The rate on those bonds is administered by the US federal government, but in fact a lot of institutions work this way, normally as a way to gain customers. A bank or credit union can lose a little bit of money by paying above-market rates if their new customers also move in their lower-rate deposits or, even better, take out higher-interest loans like credit cards. This type of account is often marketed as a “Rewards” or “Kasasa” checking account, and I keep an eye on them at depositaccounts.com (although I’m sure there are other sources; I have no relationship with that website, I just use it).

Robinhood offers bad market rates, not good administered rates

In my post on Robinhood Gold I said, “The 5% APY offered on balances up to a million or so dollars of insured deposits[…]is competitive, but it’s not best-in-class; Vanguard is paying 5.28% on uninvested cash in their own brokerage’s sweep account as of today.”

But while Robinhood was advertising the Gold product as offering 5% APY, that was not true. It was just advertising its current below-market interest rate, and when interest rates went down, its below-market interest rate also went down.

This has happened twice since my post, and Robinhood has continued to misleadingly advertise the product in this way: first “earn 4.75% APY on your cash,” and now “earn 4.25% APY on your cash.”

Vanguard’s default federal money market settlement fund has a 4.58% 7-day SEC yield as of November 22, 2024. In other words, Robinhood started out earning less than Vanguard, and instead of becoming more attractive as an administered rate it has rushed ahead of them to become less attractive as a (below-)market rate.

The Robinhood Gold fee is administered

Hopefully the problem is now coming into focus: Robinhood charges a flat $5 monthly fee, but pays a fluctuating rate on uninvested cash. At a 5% APY, it takes $2,400 in uninvested cash to break even with the monthly fee. At 4.25%, it takes $2,824.

As interest rates fall and fees stay steady, it takes more and more uninvested cash to merely cover the fees on the account, before it even makes sense to start comparing interest rates.

You cannot hold both uninvested cash and use margin

The final issue, and why I didn’t bother mentioning it in my March post, is that you cannot take advantage of the free $1,000 in margin offered as a benefit of the account while also taking advantage of the interest rate offered a benefit of the account.

You can think of this in two different ways:

• the first dollar of a stock or ETF you buy always comes out of your uninvested cash;

• and the first dollar you receive when selling a stock or ETF is always used to repay your margin first before it begins to earn any interest.

Conclusion: don’t sign up for Robinhood Gold for any of their advertised benefits

I like to think that what makes this site different is that I always try to be as honest, straightforward, and accurate as possible, so lest any of my beloved readers get misled: I plan to continue paying for Robinhood Gold.

I’ve been experimenting with Robinhood extensively and have found some specific uses (unrelated to investing) which I’m not able to describe in detail right now, but which I’m happy to pay $120 per year to continue experimenting with.

What I certainly won’t be using it for is to hold my uninvested cash: cash is too valuable to hold with people whose business model is to make sure you get as little value for your money as possible.

As for investing with the account, I’ll be using $1,000 in cash and $1,000 in margin to partially or wholly offset the annual fee, as well as hold the free stocks they give me when people (very) occasionally use my personal referral link.

Another improvement, and giving up on Venti, the gimmicky travel savings account

I’ve written before about signing up for Venti and about some improvements they made after reading my initial post. To recap, you deposit money through them with a partner bank, and you earn a notional interest rate on your deposit of 9% APY (the earning rate used to vary by account type, but they seem to have suspended that for now). The “interest” is credited as “points,” which can be used through through their booking portal to pay for part of your flight and hotel reservations.

I recently finished withdrawing my cash and redeeming the last of my points through Venti, and don’t plan on adding any more. I’ll explain why in a moment, but first I want to mention an additional feature they added recently.

Topping up cash interest with points

On July 12, 2024, I got an e-mail announcing Venti was partnering with credit unions to offer points on top of the cash interest earned on your self-managed credit union accounts, in addition to the points you earn on your “Venti Classic” balance. They call this new “cash-and-points” earning option “Venti Pro.”

As a reminder, your Venti Classic balance is held at Veridian Credit Union, but can only be managed through the Venti interface: you’re not given routing information to make deposits or withdrawals, and in fact you’re not given any information about “your” account at all. That balance earns 9% APY in Venti points, which can only be redeeemd through their hotel and airline booking portals.

Venti Pro allows you to earn Venti points on up to $25,000 in savings balances on your external accounts at their partner credit unions. They currently have four such partners:

• Wings Credit Union

• Meriwest Credit Union

• MSU Federal Credit Union

• and Georgetown University Alumni and Student Federal Credit Union.

You continue to earn cash interest as usual on those externally held accounts. But by linking them to your Venti account (through one of the usual third-party services), you’ll also earn 3% APY in Venti points on your balances up to $25,000. The program is sparse on details, so there’s no indication of how many linked accounts you can earn Venti points on, another of the many oversights Venti has shown since they launched.

The page also includes this tortured sentence: “This promotional offer is limited to new credit union accounts created within the last 30 days of your Venti account.” I’m sure this makes sense in the original Estonian, but I can make no sense of it in English.

This is such an obvious extension of Venti’s original business model that I assume it was part of the plan all along and they have been working out the kinks, either on the business or technology side. Just like with Venti Classic, credit unions pay Venti to harvest deposits for them. Venti then divides that payment by the (lower) amount they value Venti points at on their books, and turns the result over to their customer in points.

To illustrate this with some sample numbers, if in a Venti Classic account Veridian pays Venti 3% on an uncapped balance, and Venti values their points at one third of a cent each, they pay customers 9% APY on the balances they manage.

If Venti Pro credit union partners pay Venti 1% on new balances up to $25,000, then the same transformation results in the 3% APY they pay on Venti Pro-linked balances. This is surely also the reason for the tortured sentence I mentioned above: Venti Pro partners only want to pay the finder’s fee on new balances; they don’t want to pay another 1% in interest fees on existing accounts!

The MSU Federal Credit Union only pays the advertised rate on the first $999.99 in savings, and I can’t find the avertised GUAS FCU savings rate at all, but the Wings Credit Union savings account is nationally available (with a $5 membership fee to some non-profit). It offers 4.75% APY, with a $25,000 minimum opening balance and no interest earned if your average daily balance is below $25,000. The final option, Meriwest, offers 5.5% APY on the first $10,000 of your Premier Savings balance, but enforces its geographical requirements (in my experience), so is probably most interesting to folks who live in Northern California or Pima County, Arizona.

Is it worth opening a Wings account to earn additional Venti Pro points? My answer is a qualified yes: it is if you want to deposit exactly $25,000 and value Venti points at or close to their nominal value of $1 each. 4.75% APY in cash and 3% APY in “travel funds” is a great return on $25,000 in self-managed, insured cash.

But it’s not for me.

Goodbye to all that

Perhaps the most essential characteristic of a travel hacker is being game, and I’m game for just about anything. I once took the train to Philadelphia to open a prepaid debit card at a check-cashing place to earn 5% APY on the linked savings accounts (remember, interest rates were 0% for close to a decade). But when you’re game for anything, you also have to be unusually alert for warning signs.

I’ve mentioned various warning signs about Venti that were flashing yellow from the start: the slim-to-nonexistent documentation and the inconsistent descriptions of the various products did not make me terribly optimistic about the product or its long-term future.

But after all the warning signs, my red light only came on during my first Venti redemption, when I booked a flight deliciously close to the $250 point-redemption level (you can use points to pay for the first $250 of flight reservations). I booked a $258.20 flight, paying $250 with Venti points and $8.20 with my credit card (an option they added after my first post).

As soon as the flight populated to my American Airlines account, I saw that I had been booked into Basic Economy, even thought the checkout page and confirmation e-mail only said my ticket was in Economy. Since I wasn’t sure about the dates of the flight, and needed to maintain flexibility, I canceled the flight immediately through my American Airlines account. Since I’d booked the flight just minutes before, it was obviously eligible for a refund, and sure enough the $8.20 was refunded to my credit card immediately. Venti was another matter.

First, a confused Markus (who I assume runs the company, since he’s the only person I’ve ever interacted with) asked whether I had canceled my flight. I thought this was a nice personal touch, and assured him I had and mentioned why (being unable to identify a Main Cabin flight).

He replied and explained that “It skips because our broker does not provide that step for one-way flights.” Interesting, but none of my business.

He then replied a few days later and assured me that it was my user error, since he thought the website made it clear the reservation was in Basic Economy. Again, agree to disagree, none of my business.

But then Venti didn’t refund my points, which made it my business. So, I pulled my money out and redeemed the last of my points. I’m not going to war over $250 in travel credit, but if $250 is worth $83 to them (in the illustration above), it’s worth $0 to me if I can’t refund a refundable ticket, and interest rates are too high to earn 0%.

Conclusion

I’ve strived while writing about Venti to be gracious to a fault. A group of entrepreneurs struggling with English started an American company in one of the most regulated sectors of the economy to use technology to squeeze some arbitrage out of the banking system in a somewhat novel way (although it is in some ways patterned on the much-closer debit card relationships between Delta and Suntrust, Alaska and Bank of America, and American and UFB Direct).

And after all this, I still do not think that Venti is a scam. I think they really do deposit your funds with Veridian Credit Union. I think deposits really are federally insured up to the relevant maxima. But banking is an industry that is built on trust, and when you run out of trust, you run out of money pretty quickly afterwards.

Further reading:

• What Happens When Your Bank Isn’t Really a Bank and Your Money Disappears?

• Your Online Lender May Not Be a Bank. Here’s How to Keep Your Money Safe.

Quick hit: mobile balance loads trigger refreshed Amex personal Gold credits

I applied for an American Express Gold card back in April so have been following especially closely the recently announced changes to the card: a $325 annual fee, $50 semiannual Resy credit, and $7 monthly credit at Dunkin Donuts.

I already knew that purchases at any Resy restaurant would trigger the semiannual statement credit because of my experiments with my American Express Delta Business Platinum card, but I wasn’t sure whether the $7 monthly credit could be triggered without going into a Dunkin location.

So, I bought $7 of credit in the Dunkin Donuts app on Monday, July 29. On Wednesday, July 31, I received an e-mail alerting me that I’d received a $7 statement credit for the purchase.

I’m not a great consumer of either fast food coffee or fast food donuts, but as the saying goes, I’ll take any bank’s money as long as they’re giving it away. This benefit is already live, so start using it today.

Quick hit: no, Freedom family product changes don't reset your quarterly bonus spend

One of my favorite techniques is requesting product changes from Chase personal credit cards where I received a signup bonus to Freedom cards that earn 5 Ultimate Rewards points per dollar in their quarterly bonus categories, on up to $1,500 in spend per account.

This technique doesn’t get written about as much as it used to because most travel hackers prefer to chase signup bonuses, which quickly puts them over Chase’s rules for approving personal credit cards.

As a reminder, you usually can’t get a Chase signup bonus if you already have a specific card, or have received a signup bonus for it in the last 24 or 48 months, depending on the card. One way to start that clock ticking again is to request a product change to another card. My current preference is for the Freedom Flex card, since it combines the quarterly bonus categories of the Freedom (not available for new signups) and the uncapped 3 Ultimate Rewards points per dollar spent at drugstores of the Freedom Unlimited.

I had done this a number of times in the past, so I was sitting on 3 Freedom cards on which I had already reached the quarterly cap when I applied for and completed the minimum spend requirement on my new Freedom Flex. Once that signup bonus posted, I called in and requested product changes from my Freedom cards to Freedom Flex cards. Note that when doing this a new card and card number will be issued, so be sure to keep an eye on any recurring bills that are charged to your cards.

What I was unsure of was whether these product changes would reset my quarterly bonus earning cap, because I had never changed products within the Freedom quarterly bonus family; I had only moved to and from quarterly bonus cards, not between them. If product changes reset the quarterly bonus counter, then during promising quarters you could max out each card twice, as a kind of backdoor “upgrade bonus.”

But, regrettably, the quarterly bonus counter was not reset by my product change, so even though the Freedom Flex cards have new numbers, it appears everything else about the old Freedoms was saved and ported over to the new accounts.

Impressed by first MaxMyPoint hotel award alerts

A few weeks ago I read a post comparing various hotel award alert tools that concluded MaxMyPoint was the only one that did the job (since the internet is no longer searchable I can no longer find that post; if you wrote it, let me know and I’ll link to it!).

Years ago I used and recommended Seth Miller’s Hotel Hustle to find award availability optimized by point value, but the functionality of that tool died before too long and we were back to searching for award space manually.

Fortunately for me, I actually had a very specific need for an award alert, and to my shock, MaxMyPoint met it perfectly and saved me a few hundred dollars by swapping my paid reservations for award nights.

Two missing nights in Prague

The last time I visited Prague, World of Hyatt had finally secured a property in the city: the Lindner Hotel Prague Castle. To my genuine surprise and delight, it was a Category 1 property, costing between 3,500 and 6,500 points per night.

Compare this to the other obvious properties in Prague. The Hilton Prague and Hilton Prague Old Town, both great hotels I’ve stayed at many times, have standard pricing between 40,000 and 50,000 Hilton Honors points, or roughly $200-250 in value. The three centrally-located Marriott properties all cost 33,000 points and up. IHG starts in the same range: they’re non-starters.

So Hyatt giving away nights for $65 was pretty exciting. The property is located in the heart of Hřadčany flush up against Prague Castle, so while it is in a tourist area, it’s not actually an area where most tourists stay, which makes it remarkably quiet in the evening once they decamp to their hotels in the city center.

I had a simple problem: my first, third, and fifth nights were available on points, my second night was available with cash and points (for a premium room), and my fourth night was only available with cash. So that’s how I booked it: five reservations for five nights.

Then I set a MaxMyPoint alert for the entire 5-night stay, on May 17.

On May 27 at 3:22 am, I got my first alert (subject line “Your Hotel Alert Update”) saying the nights were available. Perhaps because the alert came while I was asleep, by the time I checked availability it was no longer there, if it had ever been.

On June 4, at the perfectly sensible hour of 12:08 pm, I got my second alert. This time, awake as I was, I immediately popped over to the Hyatt app and saw the alert was correct: all five nights were available. I canceled my existing 5 reservations (technically risky, but I didn’t want to transfer over another 26,500 Ultimate Rewards points if I didn’t have to) and was able to successfully complete the reservation.

Then, as if icing on the cake, MaxMyPoint sent another alert at 8:32 pm that the award space was no longer available, a charming free reminder that they had done their job for me.

Conclusion

As indicated by the history of Hotel Hustle, it’s not a great idea to let your own skill and intuition atrophy every time new tools come along to replace them, because that puts you at the mercy of the toolmaker. But it’s equally foolish to ignore when tools are introduced that you can integrate into your workflow to make your life easier, and MaxMyPoint gave me the information I needed just in time to take advantage of it. Kudos to them, and I’ll be a repeat (non-paying) customer as long as they keep it up.

Quick hit: two more free options for manufacturing debit transactions

I wrote recently about some tools I use to manufacture debit card transactions in order to trigger the highest rates on rewards checking accounts (often but not exclusively marketed under the “Kasasa” brand).

Doctor of Credit then joined in the fun, noting that one of the easiest options, adding credit to your Amazon balance, has become onerous to the point of uselessness as they raised the minimum balance reload amount to $5 from the previous $1 minimum. Unless you’re a big Amazon spender, you’ll quickly end up with more money in Amazon credit than in interest.

With that in mind, I ran a few more experiments and found two more possibilities that are working for me for now. Note that individual banks and credit unions may code transactions differently so you’ll need to verify for yourself whether these meet the transaction requirements for your accounts.

Robinhood

Robinhood, the free stock- and crypto-trading app I write about occasionally, allows you to fund your cash balance using a debit card with no fee and a minimum deposit of $1. The money is immediately available in your account to withdraw or invest, as far as I can tell, and Robinhood does support fractional share ownership so you could even use this technique to drip some of your interest into the stock market, one of many ways to exercise compounding discipline.

PayPal

PayPal also allows you to add money to your balance with debit cards, again with a minimum of $1. PayPal in principle supports multiple cards per account, so you could use a single PayPal account to meet the debit requirements on more than one high-interest accounts. However, since PayPal also doesn’t have much in the way of identity verification, if you’re considering this I would personally suggest using a new PayPal account for each debit card you plan to use, so that if one account is frozen or closed it won’t necessarily impact the others you’re using.

Conclusion

Two final quick points. First, on the account I’m currently experimenting with, both Robinhood and PayPal transactions post as “signature” transactions so should count towards my qualification requirements, but that’s something that you’ll need to monitor for each of your accounts and each of your methods. No one else can do it for you and datapoints age fast in this game.

Finally, as hinted at above, with any service you’re experimenting with to manufacture transactions you need to keep in mind two parameters: how many accounts can you have, and how many cards can you link to each account? Venmo works great for my round-up savings account because it allows transactions under $1, but I can’t link additional debit cards. The Cash app and Robinhood (with their $1 minima) only allow one debit card to be linked at a time. PayPal is more flexible on the number of cards you can have linked, but many of us have horror stories about past account closures (even though mine ultimately ended with a fat settlement check).

"How do we scale this?"

One of my favorite questions travel hackers ask is “how do we scale this?” The implicit answer is, usually, “we don’t.” Contrary to what economists pretend to believe, the world is in fact chock-full of arbitrage opportunities. What is true is that most of those opportunities are difficult or impossible to scale.

Here’s my personal breakdown of techniques used to increase scale, and the obstacles to doing so.

Brute force and constant returns to scale

The most obvious scaling strategy is to multiply your own effort. If you have a technique that generates a known amount of value per hour you spend on it, then you can get twice as much value by spending twice as much time, usually up to some limit. In a simple example, if it takes you an hour to manufacture $10,000 in in-person spend at one grocery store, and you have identical access to five grocery stores, then you can spend 5 hours and manufacture $50,000.

Automation and transformations

A lot of high-volume travel hackers focus on automation as a way to scale their techniques, and automation is one of the many tools I put in the broad category of “transformations.” Transformations are when the design of programs can be understood differently by the customer than by the business in order to scale techniques, either to get a higher return from the same amount of time or money, or to reduce the time or money needed to get the maximum return.

To give a classic example from my own practice, for many years the US Bank Flexperks Travel Rewards Visa offered 3 points per dollar spent on charity, worth up to 6% when redeemed through their travel portal. They also coded Kiva, the microlending website, as charity. People were thus able to earn as much as 6% in travel on loans of as little as a few months, and many of us did, transforming a modest discount on charitable giving into an extremely high-yield investment vehicle.

A more contemporary example is the rewards (often branded as “Kasasa”) checking accounts that offer some of the highest-earning, most liquid savings vehicles. They typically require 12-15 debit card transactions along with a direct deposit in order to earn their advertised rates. Meeting these requirements as they intend would seem to require, as they intend, reorienting your entire financial life around doing so. But when you’ve broken down the requirements to their individual parts, you can transform meeting them into a matter of a few minutes per month.

American Express cards have acquired a reputation of being “coupon books,” but a lot of the pain of redeeming those coupons (and getting back the value of your annual fee) can be transformed into painless routines:

• American Express Hilton Honors Surpass cards can be used to buy $50 in physical gift cards on the first day of each quarter to trigger the quarterly statement credit;

• Monthly statement credits for Resy and rideshare services or Shake Shack and Grubhub can be triggered by gift cards or subscription services.

All these are transformations: the company wants them to dominate your thoughts, but a few simple calendar reminders can guarantee you maximize the value of each credit without having to keep track of any of them individually.

Teammates, comparative advantage, and the benefits of trade

I call teammates everyone you partner with in order to take advantage of different circumstances, what economists call your “comparative advantage.” These can take all sorts of forms: some people have access to grocery store manufactured spend while other people have access to gas stations. Some people have more Chase cards than they’ll ever be able to maximize the value of, while others pile into American Express cards and are blocked from signing up for new Chase cards.

A lot of bloggers have a kind of “view from nowhere,” where every person has access to every credit card and each can follow prescribed steps from scratch, but it takes almost no experience to know that’s absurd. Every individual travel hacker’s situation is different, and it takes only a little more experience to identify which parts of the game you’re interested in pursuing most intensively. Finding other people with complementary interests is a way to scale each of your efforts by getting the most value from the parts of the game you’re most interested in.

The most obvious candidates for teammates are family members, precisely because there’s usually not any need to “divide” effort or results at all: everyone gets to go on the family trip, regardless of whether they made a “fair” contribution to paying for it at all. Some bloggers have affected to call these teammates “Player 1,” “Player 2,” and so on.

Employees

One of the most common questions people ask when they find out about the existence of travel hacking is, “that sounds great, but I don’t want to do it, can I just pay you to do it for me?”

There are people and situations that make this possible, but fewer than people wish or expect. The main problem is that almost anything you can train people to do on your behalf, they can do on their own behalf. You are the middle man, and unless you have both knowledge and money that are impossible to steal, your employee will quickly get the drift and go to work for themself.

The Verge had a humorous story about this very phenomenon in my home state of Montana, where resellers would continuously set up drop-shipping warehouses only to find their employees, having mastered the skill of packing and unpacking Amazon shipments after a few months simply set up their own tax-free reselling businesses.

Readers as force-multipliers

Another way to scale a technique is simply to share it. This has all the advantages of the techniques above.

Brute force techniques will have more people applying more brute force and yielding more benefits.

If you know how to transform a technique from difficult to hard, then more people will save more time and effort.

If you know how to trade personal or regional advantages with other people, then telling them how will result in more benefits for everyone in those situations.

And if you tell people how to hire employees to solve their problems, then more people will have their problems solved and more people will be employed.

The problem, of course, is that you don’t get a cut. What’s up to you is how big of a problem that is.