Hacker News Reader

Comment by danpalmer | Hacker News Reader

I recently called out my bank for advertising a "0.10%" savings interest rate, a similar problem I think.

To anyone here it's obvious that this is the same as a "0.1%" rate, but I do suspect that to a not-insignificant proportion of the population, they'd consider "10 > 1", and conclude that this account is better for them than another one offering "0.2%" interest.

I realise this might be a stretch, but to look at it another way, should a bank be allowed to advertise their savings rate as "0.100%" or "0.10000%"?

View on HN

This might be Regulation DD, set by the Federal Reserve if you're in the US, or something similar if you're elsewhere: https://www.federalreserve.gov/boarddocs/caletters/2009/0914...

> It must round the annual percentage yield, the annual percentage yield earned, and the interest rate to the nearest one-hundredth of one percentage point (.01%) and express them to two decimal places. (p.17)

Personally, I don't see what the problem would be with 0.10%, doesn't seem misleading in the slightest. I also love how the regulation is called "Truth in Savings", like how could anybody be against it?

The bank is in the UK. I don't know if there's any legislation around this communication, but it doesn't seem that great even if it's legal.

You're worried about unfair competition with banks advertising "0.1%" or "0.2%", but with a must rule like this, there wouldn't be any of those competitors. Anyone with the same rate would advertise "0.10%", and anyone who wanted to beat would say "0.20%" (or more likely, "0.11%").

Probably - iirc FSA regulate that stuff and can be quite pernickety about what they'll allow.

This sounded silly until I remembered that software versioning trips me up just the same.

I always read 1.2.11 as < 1.2.2

This is so true. The other day I spent a good 10 minutes trying to upgrade a gem to > 2.3 when it was already at 2.11

Yes, but #.#.# is not a valid number/decimal/monetary notation. You cannot enter that number into a calculator, and I'm not sure you'd even be able to use that as a valid numerical value in any programming language.

For anyone getting confused that 0.1 == 0.10, then how is that the writer's fault? There are formatting requirements that require a significant digit number of places (scientific notation as an example). I don't know that actually applies to bank marketing material, but it could be a style guide type of thing to keep numbers formatted the same. Either way, it does not change the value of the number. If you misread it, it is not the company's fault. They have provided factual information.

> If you misread it, it is not the company's fault. They have provided factual information.

That’s really not a fair argument given advertising and marketing theory and schemes.

Who knows about this particular case since we don’t know if the 0.10% needed to be compared to other rates like 0.08%, where the formatting would maybe make more sense. But it’s not beyond imagination that someone could try to hook a customer with numbers. It’s done all the time with the whole $8.99 thing.

That's funny timing, as I just mentioned the other day on here that the $.99 vs $1.00 thing was a marketing tactic. However, someone countered that it is a remnant from the old days of forcing the clerk to ring out the customer at the register to make correct change. This was to help avoid the temptation of pocketing the exact dollar amount. A pricing practice that has essentially just stuck around, and possibly repurposed????

Either way, I've learned basic maths. .1 == .10. I would be willing to guess that the average public would not think that .10 > .1, and that this whole confusion is really only for developers that have started using decimal notation in versions where .10 > .1 is true.

As an engineer, the trailing 0 is significant: it indicates with what precision it was rounded to. [0.5, .15) could become .1, with 1 significant digit, but not .10 with 2 significant digits.

Doubtful it matters in this case, but that extra zero can convey extra meaning.

Rock climbing's ratings with Yosemite Decimal System have 5.10 harder than 5.9 [1]

[1] https://en.wikipedia.org/wiki/Yosemite_Decimal_System

For this reason I tend to prefer the way OpenBSD versions their releases.

They just did 6.9 and the next one will be 7.0.

But this really only works over long periods of time because they only do two releases a year.

It took me a bit to get my head around Linux kernel versioning where the 2nd digit just goes up until they feel like it's time for a new first digit version.

In that case the change of major version loses its meaning. They could use 69 and 70.

I find normal version numbers pretty simple as a software developer (general population seemed to be buffled by how macOS 10.9 could ever be followed by macOS 10.10) but I always get confused when I see Cyberpunk 2077's version numbers that go 1.22 → 1.23 → 1.3 → 1.31. It just doesn't compute that you go from minor version 23 to minor version 3.

In the financial world we think in bps (units of 0.01%), so it would be standard to write percentages with 2 decimals so you can easily see it's 10bps at a glance.

Oh my!

I’ve often seen rates given by banks with at least one extra digit and wondered why.

It never occurred to me it was a marketing ploy.

In finance we use basis points, or bps, which is %/100. This isn't unusual.

I suspect this may be more attributable to ignorance/error than to malice/marketing, but it's still not great that no one spotted it.

> I realise this might be a stretch, but to look at it another way, should a bank be allowed to advertise their savings rate as "0.100%" or "0.10000%"?

Yes. Lotteries still exist. What's another tax on the mathematically illiterate?

I would suggest that this is a similar problem, but for a different reason. I'm not choosing a bank because they offer a 0.1%, 0.10% or 0.2% savings interest rate. That is pretty inconsequential.

How did the bank respond? Did they change anything?

"Thank you for bringing this to our attention"

That was maybe a month ago, still the same.

I'm not the most impressed, I found a bug in their iOS app that would multiply monetary values by 100 when doing various (normal) text editing steps as you input values. I found this by nearly transferring £1000 instead of £10 to a friend. The bug was an honest mistake with `NSNumberFormatter` in the iOS SDK, but I feel like a bank can't make "honest mistakes" at that sort of level.

Reply on news.ycombinator.com