Recency and attribution bias. Look over the world and over time.
Thought Bubbles from the Deep
- Thread starter DeepState
- Start date
Recency and attribution bias. Look over the world and over time.
tech/a
No Ordinary Duck
- Joined
- 14 October 2004
- Posts
- 20,417
- Reactions
- 6,356
I guess if we are clever enough we can successfully argue black is white.
( Evidently you can mathematically argue this. )
- Joined
- 25 July 2010
- Posts
- 1,120
- Reactions
- 284
My guess is that 60 years isn't enough.
Have a read of Ray Dalio's take of the Long Term Debt Cycle (I'm sure others have argued this point before, I just think Dalio does it well).
Not to say it'll necessarily follow that pattern, but an alternate view is always worth listening to (within reason)
Let me ask you a different way then tech.
Last 60y:
* Women join the workforce en masse.
* Energy production goes through the roof with low EREOI energy.
* The internet.
* Huge growth of credit economy and financial sector.
Therefore the last 60y have been essentially a giant uptrend. At least, do you concede that it is possible that this has coloured your view?
What do you think it would take for the next 60y to be like the last 60y?
Internet 2? Banks 3.0? Increase retirement age to 90 and reduce the age you can join the workforce to 12? Where is the low EREOI energy coming from? How much bigger can the financial sector get? How much larger can the debt load that supercharged the global economy go?
tech/a
No Ordinary Duck
- Joined
- 14 October 2004
- Posts
- 20,417
- Reactions
- 6,356
Don't know about coloured---I agree that NOW if you want to buy a house for profit its simply not an option.
Development is a different story though.
I agree you wont get the sort of growth that we have had.
We will get different ways of building Think Modular pre builds that take weeks to build
Whole new opportunities.
I agree that aging population and debt will alter the known landscape once again.
One of your unknown certainties.
There will be opportunity but not as we now recognise.
Right. So this is the answer to your own question, "overvalued relative to what".
Sorted!
You are conflating two separate things here. Your ability to "add value" through access to capital, knowledge, experience. As opposed to, the valuations, long term investment merits/sustainability of house prices in Australia for those with risk exposure to the sector and adjacent sectors.
Which is the actual topic DS was asking about. Not on whether it's possible for rich old white people in Australia to get richer.
tech/a
No Ordinary Duck
- Joined
- 14 October 2004
- Posts
- 20,417
- Reactions
- 6,356
Right. So this is the answer to your own question, "overvalued relative to what".
Sorted!
Sorted---time taken to return any sort of profit is not acceptable to me so over valued relative to potential return---in my foreseeable future. I don't think that's what was meant or is an acceptable measure.
Yet I can make lots on the front end developing---don't know that it ANSWERS the question of over valued relativity.
Sorry my intelligence quotient couldn't see that.
I think even moderately well healed Black/Yellow/and well tanned people can do it.
60s not old!! Just ask an 80 yr old.
Wysiwyg
Everyone wants money
- Joined
- 8 August 2006
- Posts
- 8,428
- Reactions
- 284
Comparative property prices in other locations and comparative historical prices, long term trend of income to price ratio shows an uptrend post war, new normal as they say.
DeepState
Multi-Strategy, Quant and Fundamental
- Joined
- 30 March 2014
- Posts
- 1,615
- Reactions
- 81
This would be the line from the property bulls, those who make a buck spruiking property or otherwise derive fees from transacting in it.
That argument is a little devoid of consideration of the primary drivers of those moves and simply extrapolates these as if trees grow to the moon. For a start, debt to income...which correlates rather directly with prices to income...cannot continue to escalate indefinitely. It really began in earnest with deregulation in the 1990s. As to comparison with other world markets, the ratio of house prices to average income for 26 property markets examined by the Economist puts Australia 2nd under Belgium as of 2015. Not exactly a screaming buy indicator on this measure.
http://www.economist.com/blogs/dailychart/2011/11/global-house-prices
This argument, that property is cheap/fair, is ultimately momentum oriented, new-paradigm thinking. The new-paradigms which are required to support arguments that property returns will be similar to those of the last 50 years (even in real terms) are basically impossible as central case arguments. Not surprisingly, I don't subscribe to it. It seems neither do the regulator, the central bank, Treasury or the banks themselves. Having looked into it, I tend to favour the argument that property is a bit rich.
The saving grace is foreign dumb money flooding in, buying houses that are largely not lived in by the owner or a renter. This activity is politically unpopular from different perspectives and is being actively managed with a view to constraining it. At the moment, those efforts have yet to yield fruit. Placing reliance on this type of activity as a salve for financial stability seems rather tenuous.
FWIW, it looks like about 20-30% expensive relative to equities and debt. This doesn't mean that a pop is coming. It just means that the forward looking returns aren't special in comparison to equities and debt. Nothing stops it from becoming more expensive. In the event of a correction and these momentum, new paradigm arguments are swept aside, it is definitely possible that they are replaced with a bearish perspective that we have not encountered for over 20 years. This could see a property correction of greater than 20-30% over a short period, which is why the industry is reacting the way that it is. This does not need to be central case for it to be relevant to financial stability or the pricing of credit.
For me, the actions of the major conduits of credit are in line with what I think is reasonable in the conditions. Hence, I can appreciate where Stevens is coming from on financial stability. Property and credit have been implicated in every major financial dislocation. They are a poisonous combination when extrapolative thinking takes hold. This is clearly present in Australia. It would be a mistake to make short term real activity gains at the cost of increasing financial stability risks from these levels. This is especially the case when unemployment has peaked lower and earlier than previously thought and activity is picking up in dwelling investment.
I think the credit market is about right in balancing these arguments. A slight bias to a further cut. It also leaves the AUD with more of the forward looking workload to rebalance our economy. Wishing doesn't make it so but, so far, so good.
- Joined
- 30 August 2011
- Posts
- 88
- Reactions
- 1
Very well considered post DS. Now where is that like button.
- Joined
- 4 October 2012
- Posts
- 871
- Reactions
- 3
Should it break through the 4.44% line then a double bottom will be confirmed...
Wysiwyg
Everyone wants money
- Joined
- 8 August 2006
- Posts
- 8,428
- Reactions
- 284
Yes I fully agree. Thank you for an experienced perspective.
- Joined
- 2 June 2011
- Posts
- 5,341
- Reactions
- 242
A an ex-colleague, who I am now friends with, recently relocated to Melbourne from London. His primary reason was the weather and lifestyle but is amazed at how affordable housing is. I don't know what sort of $$$ he's pulling in but he works in a fairly senior position for a largish pension fund, so I'm not talking about someone making the median salary. In his opinion Australians don't know how easy they have it in property (his words not mine!). Of course he agrees that property is a very poor investment, but the people pushing up property in western countries aren't necessarily approaching it from an investment perspective.
DeepState
Multi-Strategy, Quant and Fundamental
- Joined
- 30 March 2014
- Posts
- 1,615
- Reactions
- 81
My view is that gold is a currency in waiting. This view is taken from the practices of the reserve managers of the largest economies and monetary institutions as well as from historical observation and the customs relating to gold in India and China.
That gold has functioned as the basis of money is not in doubt. For various reasons, alternative perspectives that gold is not a monetary asset keep arising. They appear partially driven by Gold's fall in (USD) price in the last approx. 3 years, raising a question about whether they are a monetary asset and whether it serves as a store of wealth.
Let's deal with the issue of whether gold is more like a currency or an industrial commodity in more recent times. Over longer time periods, this is not in doubt. Over the shorter term, casual observation can give rise to a belief that gold is functioning less as a monetary asset and more like an industrial commodity. This issue is capable of direct analysis.
Over the last three years, gold has performed much more similarly to major currencies than it has to Oil, Copper and Corn. Gold continues to show characteristics of a currency rather than a standard bell weather energy, industrial metal, or agricultural commodity. The call on whether gold is closer to a monetary asset or industrial commodity on the ultimate outcome of interest...which is returns...is not even close over the more recent period in which gold has lost favour (in USD).
This analysis is based on a well-accepted method to assess similarity in returns (or other features under examination). If of interest: PCA, Euclidian distance (Ward), daily data.
I might leave it to others to discuss the validity of questioning gold's properties as a store of value based on the post 2012 performance in USD.
That gold has functioned as the basis of money is not in doubt. For various reasons, alternative perspectives that gold is not a monetary asset keep arising. They appear partially driven by Gold's fall in (USD) price in the last approx. 3 years, raising a question about whether they are a monetary asset and whether it serves as a store of wealth.
Let's deal with the issue of whether gold is more like a currency or an industrial commodity in more recent times. Over longer time periods, this is not in doubt. Over the shorter term, casual observation can give rise to a belief that gold is functioning less as a monetary asset and more like an industrial commodity. This issue is capable of direct analysis.
Over the last three years, gold has performed much more similarly to major currencies than it has to Oil, Copper and Corn. Gold continues to show characteristics of a currency rather than a standard bell weather energy, industrial metal, or agricultural commodity. The call on whether gold is closer to a monetary asset or industrial commodity on the ultimate outcome of interest...which is returns...is not even close over the more recent period in which gold has lost favour (in USD).
This analysis is based on a well-accepted method to assess similarity in returns (or other features under examination). If of interest: PCA, Euclidian distance (Ward), daily data.
I might leave it to others to discuss the validity of questioning gold's properties as a store of value based on the post 2012 performance in USD.
DeepState
Multi-Strategy, Quant and Fundamental
- Joined
- 30 March 2014
- Posts
- 1,615
- Reactions
- 81
Nice.
Are these PCA done from total return? i.e. i.e. accounting for contango on the futs and carry on the FX?
DeepState
Multi-Strategy, Quant and Fundamental
- Joined
- 30 March 2014
- Posts
- 1,615
- Reactions
- 81
Standardised into Z(0,1) hence looking at correlation. Sizing can be adjusted so correlation is the focus. All figures based off spot. Hence no contango/backwardation effects on commodities.
No carry effects accounted for Spot FFX either. Nice point though. Very astute indeed. FFX would be the most affected issue even on spot but interest rate differentials on overnight cash have been basically static and thus not materially impact correlations if at all (AUD would be marginally affected). Either that or I will just say that AUD and other FFX return is for cash under the mattress. Could run this again using overnight mid prices, but it's not worth it.
The main point here being that it's pretty impossible to hold the spot in the commodities you picked and certainly impossible to trade a barrel of brent or bushel of corn in a liquid market. I think if you used front month futs and adjusted for roll yield (e.g. take a look at USO ETF on the NYSE) you can see the big difference between the spot price and what you actually can "invest" in.
(ergo the return on gold is even more like a currency and less like a commodity)
Good point about the differentials, since they haven't moved much I guess we can let it slide. I definitely try to account for it, as I have noticed that the difference in timeseries can have a material impact on backtests.
DeepState
Multi-Strategy, Quant and Fundamental
- Joined
- 30 March 2014
- Posts
- 1,615
- Reactions
- 81
Enjoying this discourse.
Had a look at USO ETF as suggested. It is fully replicating the S&P GSCI Crude Oil index. Although you will pay 0.66 MER for the pleasure of gaining this exposure via the ETF. I had used Brent in my analysis, but whatever. You make a good point that the spot market is not accessible. (Well, you can actually buy the spot contracts on IG Markets...still I found your point interesting).
I was curious about the extent to which the character of returns was distorted by basis risk. Anyhow, FYI only, the R-squared for S&P GSCI Crude Oil Index vs WTI Spot is 0.95 with correlation of 0.98 for the last 3 years based on daily data. Basically, in this case, I acknowledge that the relationships portrayed in my analysis may not be exact for the investible equivalents (if that means actually holding the stuff in your hands). However, if I had added the investible equivalents in the form of front month contracts it would produce a plot which is so overlapping that the labels would be hard to read. Basis risk is almost completely subsumed by the volatility of the underlying. I imagine this would be the case for the other commodities examined as well. Still, a good point....and hugely important for trading strategies for alpha, just less so for correlations.
When doing backtests for currency in terms of alpha generation, it is definitely important to include carry effects. However, for the purposes of identifying correlations, where interest rate differentials are not moving around a lot, the outcomes are pretty much identical because allowing for carry adjusts for the level effect...which is controlled for anyway in PCA. The carry effects also come with a spread due to the tom-next margins. Always clipping a ticket somewhere...
Hey DS, been following a bit of the discussion in the gold thread, and it's been stated quite a few times that are a few of the posters, yourself included, use gold as a kind of insurance protection for your portfolios. Whilst if I was going to calculate how much insurance I needed for my house or car, it would appear pretty easy to do. However, permanent risks to an investment portfolio are obviously much harder, or even impossible, to quantify. What sort of methods do you know that estimate how much portfolio insurance you may need?
