6 ms·
Makes you wonder what the overhead on this was compared to just giving all the bonds an interest rate. A lot of things we do today in IT feels overly expensive
by wodenokoto 7y ago
Makes you wonder what the overhead on this was compared to just giving all the bonds an interest rate.
A lot of things we do today in IT feels overly expensively done, but looking at this really makes you appreciate how far we’ve come in terms of efficiency even with modern over engineered solutions.
- refset 7y agoState-backed application of behavioural economics [0] was the driver behind the RNG overhead, not efficiency: "Premium Bonds were introduced [...] to control inflation and encourage people to save" [1] [0] https://en.wikipedia.org/wiki/Prize-linked_savings_account#Economics_of_prize_linked_savings https://en.wikipedia.org/wiki/Prize-linked_savings_account#E... [1] https://en.wikipedia.org/wiki/Premium_Bond#History https://en.wikipedia.org/wiki/Premium_Bond#History
- LeoPanthera 7y agoThe randomness of premium bonds are the whole point. If people wanted a flat interest rate they would just buy a normal bond, or some other kind of investment. Premium bonds are so popular precisely because they are unpredictable. It's like buying a lottery ticket, except unlike a lottery ticket, you get your money back if you don't win.
- pintxo 7y agoAlternatively, you could put your money in a normal bond and invest the interest payout into lottery tickets.
- m-i-l 7y agoIts a bit more hassle to do that, and I think you're likely to be worse off for your efforts. The UK 10Y Government Bond currently has a 0.525% yield[0], and the UK National Lottery prize fund is 47.5%[1], meaning in the exceptionally long term you'd turn your 0.525% return into a 0.249% return. If you factor in that the Lottery prize fund is weighted very heavily towards the larger sums that you'll realistically never win, then the likely return within a normal lifetime is probably around half of that. Compare with the Premium Bond's 1.4% return[2]. [0] http://www.worldgovernmentbonds.com/country/united-kingdom/ http://www.worldgovernmentbonds.com/country/united-kingdom/ [1] https://en.wikipedia.org/wiki/National_Lottery_(United_Kingdom) https://en.wikipedia.org/wiki/National_Lottery_(United_Kingd... [2] https://www.nsandi.com/interest-rates https://www.nsandi.com/interest-rates
- quietbritishjim 7y agoThis was very interesting thanks! I was especially surprised that the effective interest rate of Premium Bonds is 1.4%, which is pretty good compared to today's savings rates in the UK. When I last checked (quite a few years ago) Premium Bonds were quite a poor deal. It's a little unfair that you said "the Lottery prize fund is weighted very heavily towards the larger sums that you'll realistically never win" but don't qualify that the effective interest rate of Premium Bonds also includes some high value but unlikely prizes. I wonder what the interest rate is for 50th percentile prize winners. Edit: This page[1] on MoneySavingExpert shows the median (50th percentile) Premium Bond winnings are 1.16% over a year if you have £15,000 invested, and are 0%(!) if you have £1,000 invested. They also have a winnings calculator[1], which shows that median probabilities don't compound over time in the way you'd expect: for £1,000 over 5 years the median return is £50, which is equivalent to 0.98% compound annual interest. (It may actually be fractionally higher because I don't think the calculator assumes you re-invest the winnings in Premium Bonds.) [1] https://www.moneysavingexpert.com/savings/premium-bonds/ https://www.moneysavingexpert.com/savings/premium-bonds/ [2] https://www.moneysavingexpert.com/savings/premium-bonds-calculator/ https://www.moneysavingexpert.com/savings/premium-bonds-calc...
- m-i-l 7y agoThat sounds right. Using the latest figures I can find online, the lowest Premium Bond prize (£25) was £75,665,750 of the total £88,099,800 prizes[0], i.e. over 85%, whereas lowest National Lottery prize (free ticket worth £2) was £2,306,736 of the total £19,891,729 prizes[1], i.e. under 12%. So if you calculate on only ever winning the lowest prizes, the Premium Bond's not too bad notional return is likely to be not that much worse in practice, but the unfavourable distribution towards the headline grabbing large jackpots means the National Lottery's very bad notional return is likely to be significantly worse in practice. [0] https://www.nsandi-adviser.com/july-2018-premium-bonds-prizes https://www.nsandi-adviser.com/july-2018-premium-bonds-prize... [1] https://www.national-lottery.co.uk/results/lotto/draw-history/prize-breakdown/2516 https://www.national-lottery.co.uk/results/lotto/draw-histor...
- hhmc 7y ago> The randomness of premium bonds are the whole point. This isn't strictly true, there are other reasons why they might be attrative. Premium bonds are tax advantaged (as opposed to most other investment vehichles in the UK). For certain income brackets, they are a reaosnably efficient liquid store.
- wodenokoto 7y agoI know. It’s still interesting to know what the overhead of that operation is compared to normal bonds
- altacc 7y ago> looking at this really makes you appreciate how far we’ve come in terms of efficiency My exact thoughts too. It's incredible that within a few decades, a process which was a full time job for dozens of people could now be totally replaced with code that runs in seconds. Where I work we often have to involve some manual steps in business processes (too complex or edge case to develop in the time available, so we hand over to customer support or back office staff). When that happens it often feels like we're failing to do enough to support those colleagues. But I rarely stop to think that a short time ago absolutely everything they do would have been manual.