I’m giving a talk on subprimes to the “Monty Pelican Society”:
- Date: Wednesday April 2nd
- Venue: Sydney Mechanics School of Arts, 280 Pitt St, Sydney NSW 2000 (near Town Hall)
- Time: 6.30pm-8pm
- For more information, contact Troy Henderson (troyh@search.org.au), or just rock up on the night.
April 1st, 2008 at 3:20 pm
Hello Steve,
Is it possible to record the audio of the speech and put it up for download?
regards.
April 2nd, 2008 at 11:24 pm
It’s been recorded; I’ll have to wait for the organisers to get a podcast file to me before I can post it though.
April 3rd, 2008 at 7:02 am
Hi Steve!
Looking forward to your podcast.
Someone from another blog posed this question:
How would you answer this?
April 3rd, 2008 at 9:36 am
Hi Contrarian,
I hope to have two podcasts this month: one already recorded with Stu Cameron from Rife Media, and the other of a talk to the “Monty Pelican Society” last night on subprimes.
On the question, while it is true that the distribution of debt matters, so does the absolute magnitude. At its simplest level, debt and the financing obligations it generates are a monetary claim on the productivity of the physical, “real” economy.
The debt level we have now, combined with high real interest rates, translates into almost 15 percent of real output being required to meet interest payment commitments. That is an enormous burden, compared to the long term average of around 2-3%, and it’s a burden only ever experienced before during bouts of extreme deflation.
There is also a way in which a highly focused debt can be–but not necessarily is–less of a problem than widely distributed debt. The “chain reaction” effect of one bankruptcy precipitating another could be less if the debt was highly concentrated in just one fraction of the population. They would go bust, those with whom they had direct financial dealings might also go bust, but the contagion might not spread much further.
When debt is widespread however, the contagion can go a long way.
April 3rd, 2008 at 4:32 pm
The subprimes may be similar to the canary in the coalmine, in having brought attention to a lot of the other excessive debt. If debt had continued to expand for a few more years they would have had many more standard mortgages in difficulty. There actually are a lot already, only half of the foreclosures in America are subprime mortgages.
I suspect that what we have in Australia is investors relying on subprime tenants, in other words you don’t want to be relying on them to pay your mortgage for you. In a recession the undersupply of housing could easily turn into an oversupply.
April 3rd, 2008 at 9:44 pm
Hi Steve,
I was at your talk on Wednesday night. It helped put a few more pieces together in my understanding of the mess. Thanks.
During the talk you mentioned that the average yearly return from the Dow Jones was something like 12 per cent from 1915 to now. I was surprised by this as recently I was having a read of Warren Buffet’s letter to his shareholders (you can download a copy at the berkshire hathaway website). In it, he discusses the return from the Dow between 1900 and 2000. He got 5.3% or so (the Dow at the beginning of 1900 was at 66, at the end of 200 was something like 11500, giving a compound return of 5.3% p.a., have a look, the entertaining analysis starts at page 18.)
Considering that you started from 1915 and went through to now, I had a look at the numbers and find that the return would be fairly similar (I think 1915 the Dow was at 55 or so and now, its roughtly 12500 say, a return of 6%p.a or so for 93 years.) Am I doing something wrong!?
Buffet’s point being of course that people in the finance industry (and for that matter at dinner parties!) going around that telling people should expect something like 8-10%p.a. from shares or even from property is a bit rich. The fees they charge are even richer! I mean consider the Dow Jones in the year 2100. If it were to do 10%p.a. between 2000 and 2100 it would stand at something like 159,000,000!! The ASX200 would be something like 35million. Apply the same to an average house in Australia. 10%pa for 100 years, with an average price of say $350k, you get $5.5billion, yes billion! And that’s for an average house! Maybe my sums are a bit screwed, but I learned compound interest in year 7 or 8 – it’s pretty simple. 10%pa just doesn’t make sense, even 8% doesn’t make sense, especially when you consider how much debt we already have and the fact that wages aren’t going up in a hurry.
Some may argue that we live in a new paradigm – what new paradigm would that be – we borrow money and don’t have to pay it back or maybe we have the 200 or 300 year loan! It’s just crazy. At some point, people are just going to say no I’m not paying it. Mind you I’ve been saying this since 2002 and people keep paying it!
Another question which I probably should have asked at the talk was if the parties holding the CDOs are essentially pension funds and investors looking for stable cash type returns, why have the banks been writing off so much debt? Is it a case that they were caught with their pants down whilst trying to sell the CDOs. In other words, they were in the process of writing out many more CDOs just as people were finding out that the ones they were already holding were crap? I thought the banks were up to their usual tricks of passing on all the risk and taking all the cream.
Thanks again.
Regards,
Nikola.
April 3rd, 2008 at 10:31 pm
Hi Nikola,
I wish you had asked those questions! Actually, maybe not live; I had to go back to my data to see where I got that from, and I made a pretty elementary stuff-up which I doubt that I could have explained all that well on my feet.
The daily average using the data I had till then was what I said it was, but the performance of the Dow Jones is highly nonlinear: multiplying the daily average by the number of days in the year won’t give you the yearly figure. That’s the mistake I made in that extrapolation I gave you to a 12% yearly return from the fractional daily return.
In fact, with the longer time series I now have, the average daily movement is now negative! But that’s the linear average of a highly nonlinear time series.
Buffett’s exponential fit result is much more correct for the long term than I have–I’ll make sure I add an exponential fit to anything like this that I present in future.
The impossibility of sustained asset inflation above inflation returns forever that you highlight is something I teach my students by getting them to give me a desired rate of return (typically 4-6% p.a. in real terms), and then showing them what that would mean if it were sustained forever.
The illustrations are rather like yours: a 2% rate of return from the year dot to now on $1 would yield a ball of gold 1.3km in diameter; 4% gives you a ball of gold 2/3rd the diameter of the sun; and 6% I think returns a ball of gold larger than the Milky Way.
The banks are writing off debt because they also have holdings of these bonds–they too were buyers from loan syndicators, chasing “high and safe” AAA returns. They also have some on repo agreements, hence all the Central Bank rescues and their acceptance of below AAA grade securities (at a discount) for repo agreements.
April 8th, 2008 at 9:48 am
The Long term return from the DOW or S&P 500 is the combination of both the capital gain + the dividends.
The capital gain from the DOW over the period 1900-2000 (100 years) 11500/66 works out to be 5.4%. The annual dividend in 1900 was 4% and rose to 9% in 1932 and recently it has been @ 1-2%. The total 100 year return will work out to be roughly 10 – 11% compound
April 8th, 2008 at 11:53 pm
Yes Peter,
But there is survivor bias in the DOW, in two ways (the DOW is continually–though not frequently–updated on the basis of market capitalisation), and if members of the Dow haven’t actually gone bankrupt, some members of wider indices (such as the S&P500) have.
I’m not sure of the remnant after that effect is accounted for, but doubtless it would wipe a large amount off the compound return. $100 invested in the Dow companies of 1915 would certainly not have returned 10% compound for the last 90 years.
April 9th, 2008 at 1:03 pm
In large part I agree however I think that owning the index does account for survivor bias.
The two main takeaway points being made by people like Warren Buffett are 1. In aggregate investors cannot expect long term Capital Gain higher than what long term gains in reality are and Capital Gains in aggregate are directly proportional to GDP growth. 2. The start price + dividend yield and the end price + dividend yield will make a huge impact on the periods total returns. Warren Buffett was recently predicting roughly 7-8% total return and he believes this based on the present higher than trend normal purchase prices and lower than trend normal dividend yields (Higher purchase prices imply lower yields and lower total returns)
April 9th, 2008 at 3:27 pm
A perspective for so called ‘investors’ who relatively recently over borrowed and over paid to buy assets (houses and stocks) at (by historical measures) inflated prices. An interesting calculation to perform is Capital Growth + Yield – Cost of debt. For a ‘house investor’ the present long term (100 year) price growth at best will be roughly (GDP) 5% + (the yield) 3% – 9.25% (debt cost) = – 1.25% total return. The answer for the stock market is similar. If prices regress to more historical (100 year) levels over the next 10 years, their 10 year results will be huge paper losses for debt financed ‘investors’ (tongue in cheek – speculators). The probability that this may occur could be why Warren Buffett (who did not become the richest guy on the planet through faulty analysis and logic) is sitting on USD $65 Billion of cash (and cash equivalents) and it appears he’s in no great hurry to ‘invest it’ at present prices.
April 10th, 2008 at 1:15 am
Steve
Unless I misunderstand the source material from Crestmont Research…
The S&P 500 index was 6.1 in 1900 with roughly a 4% dividend yield.
In 1900 the p/e was 18.6 in 2000 the p/e was 41.7
(The p/e roughly doubled over the 100 years 1900 – 2000 however it is almost the same in 2008)
The 100 year capital return for the S&P 500 index was roughly 1427/6.1 = 233 times over 100 years = roughly 5.6% = roughly nominal US GDP growth, and the dividends 4 – 5%
The 100 year average S&P 500 index dividend was 4.4%
The total S&P 500 index return has been roughly a nominal 9 – 10% p.a.
The 100 year average nominal US GDP was 6.6%
The 100 year average US inflation rate was 3.3%
November 13th, 2008 at 1:08 pm
zmgkqjhh6gr09as8