Annual letter from Prem

[Jurgis]

MSFT had a moat for 20+ years (maybe 30, depends how you count).

 

GOOGL is likely to have a moat for 20+ years (but ~10-15 already have passed).

 

Some of the non-tech "moats" have collapsed in 20 years.

[barsax]

On March 4th, the day after the equity issue closed, CIBC resumed coverage on FFH, rating them a sector under-performer with a target price of $475.  What were they thinking??  Of course, the stock is just under an all-time high, and the Brit acquisition (Q3 close) will bring FFH investment portfolio up to about 30-32 billion $USD.  Decent investment returns, combined with improved insurance results should produce some pretty hefty profits throughout 2015...looking forward to Friday's PPI numbers.  I understand China's PPI numbers were pretty soft last night.

[petec]

I don't really understand why people are so negative on the equity hedges.  Sure, the headline cost is big, but:

 

1) These guys are not asset managers and should not be measured as such.  They are insurers who have to answer to regulators and ratings agencies and clients, and they invest most of the shareholder equity in equities.  That means they can lose the company if they get the equity portion wrong.  They saw market risks (and let's not forget there were genuine market risks) and hedged themselves.  Effectively, they ensured that they could not lose the company while keeping exposure to any alpha they could generate (the real problem is that I don't think their equities outperformed).  I hope they continue to shepherd the NPV of the insurance company they have built so wisely.

 

2) I also think it is worth noting that they hedged only listed equities.  They also poured money into unlisted equities (like Zenith) which they did not hedge.  In other words they simply removed any possibility that regulators, rating agencies, and clients could harm them in the event of a market panic.  They took a lot of equity risk, just not market risk.

 

3) Yes, I'd be richer now if they hadn't hedged.  But I didn't have to own the stock.  They told us what they were doing.  Anyone who owned the stock through the hedged period and now complains about the hedges hasn't got a leg to stand on imho.  Credit to those who timed their ownership better, but FFH has helped me sleep at night so I'm not complaining.

 

4) It has been but a few years.  I know we are all trained to think 5 years is a loooooong time in investing but it isn't.  Big leverage cycles work out over far longer than that.

 

P

 

 

[petec]

On the deflation front, I'm with Gio.  Prem has called a commodity collapse very well but I don't think that's the root of his deflation thesis.  He is worried about a possible deleveraging-driven deflation, which could/will happen when the capacity of borrowers to borrow is exhausted and asset prices start declining.  That's why he keeps referencing Japan and the 1930s, and has said that reflating the global economy is going to be very difficult.  So far I believe the evidence is in his favour: massive worldwide money printing has produced rising global debt/gdp, which has driven asset inflation, but there is precious little growth to support either.  In fact, if anything, the excess capacity driven by all those new assets (factories in China especially) is deflationary.  Who knows what the future holds but it has proved very difficult to reflate the global economy, and if asset prices do start falling, as Gio says, it could well trigger something very unpleasant.

[Packer16]

To tell the truth I think Prem is wrong on deflation at the CPI level and the data has and will continue to show this.  At the same time after bubbles bursting in the US in the 1930s and Japan, both were in deflationary spirals and this has not happened due to the monetary QEs.  This is how Sweden was able to blunt the effects of the depression on itself.  There is no incentive to increase interest rates (which would usher in Prem's scenario - i.e. high real rates) until inflation occurs.  I just do not see how (absent a war or some other event which will permanently destroy large amounts of capital) that interest rates will go up when there is a huge demand for fixed income securities.  Look what happened when the last QE was stopped: LT interest rates went up then back down again to where they were when QE was going on.  The only reason we have deflation so far at the CPI level is due to commodity (energy) deflation which I think is more of a one-time versus recurring event.  Just my view.

 

Packer 

[giofranchi]

Packer,

 

I don’t understand: in my mind low interest rates are a symptom of the market forecasting deflation to come…

 

I also don’t understand why you keep pointing at Sweden in the 30s’: Sweden back then was a minuscule economy which simply was successful in exporting deflation… It probably had been allowed to do so given its very small size… But what does an economy of just a few million people have to do with the situation of the US, Europe, China, and Japan today, which are trying to deleverage simultaneously?

 

Gio

 

[Packer16]

There is another reason for low interest rates: excess capital.  Historically capital has been destroyed in wars and with disease and famines.  These have not happened on scale since WWII so with the falling of the communist system and the development of wealth by most countries around the world, you have a large amount of capital generation with little destruction.  You also have a turbocharge of the developed world having a safe haven for wealth created in the developing world.  The last time this happened was between the Napoleonic Wars and WWI.  This was a time of low inflation and deflation.  I agree that there will be real deflation but the question for the CPI hedges is will there be nominal deflation.  If we were on a gold standard or had stable currency regime (like Switzerland), there would be deflation but we are not.  The only time we observed CPI inflation was when commodity priced collapsed but the core chugged along at 1.5% per year.  Some say wait and you will see deflation but we have been waiting since 2008.  My question is how much longer do we have to wait for core CPI inflation or under what conditions will the deflation theory be disproved?

 

Sweden is the only example we have of QE in a deflationary spiral in history that actually worked to stop deflation.  You can quibble with the applicability but I know of no other examples of QEs application in history, do you?

 

Packer   

[mcliu]

Sorry to interrupt the conversation, but what's real inflation and nominal inflation?  :o

[Cardboard]

The other problem with the CPI index is that governments have been specialists at including items in that basket of questionable value to reduce readings of inflation over time. This was keeping people quiet in terms of demanding wage and pension increases.

 

What if they do the opposite going forward to prevent people from thinking that goods will get cheaper in the future and to prevent the economy from tumbling?

 

Cardboard

[benhacker]

What if they do the opposite going forward to prevent people from thinking that goods will get cheaper in the future and to prevent the economy from tumbling?

 

You think regular folks look to the CPI to make their purchase decisions?  I personally don't think so... most people I talk to always seem to think their "cpi" is higher than official (probably because CPI aggregate is lower than cpi equivalent for bottom 2/3rds of income).  If the government starts increasing CPI artificially, I would guess people would not notice or think the government is sandbagging which would make them fear inflation more.

 

Just my 2 cents.  I kind of feel like Packer about Fairfax's CPI bets... but you never know.  Maybe it's just a play on a world war without saying as much... don't know.

[giofranchi]

There is another reason for low interest rates: excess capital.  Historically capital has been destroyed in wars and with disease and famines.  These have not happened on scale since WWII so with the falling of the communist system and the development of wealth by most countries around the world, you have a large amount of capital generation with little destruction.  You also have a turbocharge of the developed world having a safe haven for wealth created in the developing world.  The last time this happened was between the Napoleonic Wars and WWI.  This was a time of low inflation and deflation.  I agree that there will be real deflation but the question for the CPI hedges is will there be nominal deflation.  If we were on a gold standard or had stable currency regime (like Switzerland), there would be deflation but we are not.  The only time we observed CPI inflation was when commodity priced collapsed but the core chugged along at 1.5% per year.  Some say wait and you will see deflation but we have been waiting since 2008.  My question is how much longer do we have to wait for core CPI inflation or under what conditions will the deflation theory be disproved?

 

Sweden is the only example we have of QE in a deflationary spiral in history that actually worked to stop deflation.  You can quibble with the applicability but I know of no other examples of QEs application in history, do you?

 

Packer 

 

Imo extremely low interests rates are engineered by central banks on behalf of governments around the world, simply because they cannot afford to pay higher interests on their debts. Period.

And this might go on until prices stay very low or else go down. Higher interests rates will exacerbate the debt problem, therefore rendering the final outcome even more unpredictable and dangerous.

 

Watsa has always said that in Japan the printing of a lot of money stopped deflation for 5 years after their stock market and real estate bubble burst… The US probably has printed even more money than Japan did! So I guess we still have to wait at least two or three years… And of course the next serious market correction! If financial assets start to go south, that would be very deflationary imo! After that, if we still have no deflation, I would simply let those CPI contracts expire.

 

I don’t understand how the fact Sweden is the only example available means it should be relevant… But probably it is only me! :)

 

Gio

 

[tombgrt]

There is another reason for low interest rates: excess capital.  Historically capital has been destroyed in wars and with disease and famines.  These have not happened on scale since WWII so with the falling of the communist system and the development of wealth by most countries around the world, you have a large amount of capital generation with little destruction.  You also have a turbocharge of the developed world having a safe haven for wealth created in the developing world.  The last time this happened was between the Napoleonic Wars and WWI.  This was a time of low inflation and deflation.  I agree that there will be real deflation but the question for the CPI hedges is will there be nominal deflation.  If we were on a gold standard or had stable currency regime (like Switzerland), there would be deflation but we are not.  The only time we observed CPI inflation was when commodity priced collapsed but the core chugged along at 1.5% per year.  Some say wait and you will see deflation but we have been waiting since 2008.  My question is how much longer do we have to wait for core CPI inflation or under what conditions will the deflation theory be disproved?

 

Sweden is the only example we have of QE in a deflationary spiral in history that actually worked to stop deflation.  You can quibble with the applicability but I know of no other examples of QEs application in history, do you?

 

Packer 

 

Imo extremely low interests rates are engineered by central banks on behalf of governments around the world, simply because they cannot afford to pay higher interests on their debts. Period.

And this might go on until prices stay very low or else go down. Higher interests rates will exacerbate the debt problem, therefore rendering the final outcome even more unpredictable and dangerous.

 

Watsa has always said that in Japan the printing of a lot of money stopped deflation for 5 years after their stock market and real estate bubble burst… The US probably has printed even more money than Japan did! So I guess we still have to wait at least two or three years… And of course the next serious market correction! If financial assets start to go south, that would be very deflationary imo! After that, if we still have no deflation, I would simply let those CPI contracts expire.

 

I don’t understand how the fact Sweden is the only example available means it should be relevant… But probably it is only me! :)

 

Gio

 

 

But I guess Watsa can point at a single example to make his case? What's the difference?

 

 

[petec]

To tell the truth I think Prem is wrong on deflation at the CPI level and the data has and will continue to show this.  At the same time after bubbles bursting in the US in the 1930s and Japan, both were in deflationary spirals and this has not happened due to the monetary QEs.  This is how Sweden was able to blunt the effects of the depression on itself.  There is no incentive to increase interest rates (which would usher in Prem's scenario - i.e. high real rates) until inflation occurs.  I just do not see how (absent a war or some other event which will permanently destroy large amounts of capital) that interest rates will go up when there is a huge demand for fixed income securities.  Look what happened when the last QE was stopped: LT interest rates went up then back down again to where they were when QE was going on.  The only reason we have deflation so far at the CPI level is due to commodity (energy) deflation which I think is more of a one-time versus recurring event.  Just my view.

 

Packer

 

I also have a nagging feeling that deflation will never make it into CPI but there is a clear mechanism for it to do so:

 

1. QE works by incentivising debt, which either draws forward demand or is used to build productive assets.  That sets up a future period where demand is slack and there are a lot of productive assets. That is clearly deflationary and explains PPI deflation in China today and the US in the 30s. 

 

2. QE also works by inflating asset prices to create a sense of wealth, which "ought" to feed into demand.  I say "ought" because it also makes buying a house and saving for retirement more expensive, so I suspect it actually doesn't increase demand all that much.  But if you do believe that QE is inflationary while it is happening, then surely it is deflationary when it stops?

 

Unless, of course, it doesn't stop. and every bout of deflation is met with money printing.  Then you will eventually see something very unpleasant: a loss of confidence in the currency.  That is what causes people to sell fixed income securities in a world of excess capital: people dump bonds to buy stuff that will hold its real value, not just its nominal value.  At that point, central banks levers are no longer effective in the economy, because they work through a currency that no-one wants any more.  I do worry that FFH is not well protected against this possibility but it also seems distant at the moment.

 

To summarise: QE might be very deflationary (it encourages debt and bubbles and misallocation of capital, all of which are deflationary) and it might be inflationary (if it leads to a loss of confidence in the currency).  It is also possible that QE is just the tonic the economy needs, and if judiciously administered will get us to escape velocity and then be withdrawn.  I hope so but I'm afraid I remain sceptical.

 

P

 

[petec]

There is another reason for low interest rates: excess capital.  Historically capital has been destroyed in wars and with disease and famines.  These have not happened on scale since WWII so with the falling of the communist system and the development of wealth by most countries around the world, you have a large amount of capital generation with little destruction.  You also have a turbocharge of the developed world having a safe haven for wealth created in the developing world.  The last time this happened was between the Napoleonic Wars and WWI.  This was a time of low inflation and deflation.  I agree that there will be real deflation but the question for the CPI hedges is will there be nominal deflation.  If we were on a gold standard or had stable currency regime (like Switzerland), there would be deflation but we are not.  The only time we observed CPI inflation was when commodity priced collapsed but the core chugged along at 1.5% per year.  Some say wait and you will see deflation but we have been waiting since 2008.  My question is how much longer do we have to wait for core CPI inflation or under what conditions will the deflation theory be disproved?

 

Sweden is the only example we have of QE in a deflationary spiral in history that actually worked to stop deflation.  You can quibble with the applicability but I know of no other examples of QEs application in history, do you?

 

Packer 

 

Imo extremely low interests rates are engineered by central banks on behalf of governments around the world, simply because they cannot afford to pay higher interests on their debts. Period.

And this might go on until prices stay very low or else go down. Higher interests rates will exacerbate the debt problem, therefore rendering the final outcome even more unpredictable and dangerous.

 

Watsa has always said that in Japan the printing of a lot of money stopped deflation for 5 years after their stock market and real estate bubble burst… The US probably has printed even more money than Japan did! So I guess we still have to wait at least two or three years… And of course the next serious market correction! If financial assets start to go south, that would be very deflationary imo! After that, if we still have no deflation, I would simply let those CPI contracts expire.

 

I don’t understand how the fact Sweden is the only example available means it should be relevant… But probably it is only me! :)

 

Gio

 

 

But I guess Watsa can point at a single example to make his case? What's the difference?

 

He regularly points at two.  But personally I think there's relatively little use looking at any of these examples except for context.  Every economy and policy is different.  There's evidence that QE has worked in the US and evidence that it is not working in Japan (other people would probably say the opposite).  There's a good theoretical argument that what China needs is a tighter monetary policy, which is counterintuitive.  It's useful to know that QE and its equivalents helped in Sweden (if it did - I don't know anything about that).  But it's also useful to know that it's failed elsewhere and caused hyperinflation in yet more places.  Analysis needs to be a lot more complete than just pointing to one example (no offence to Packer or Prem!).

[giofranchi]

But I guess Watsa can point at a single example to make his case? What's the difference?

 

I was simply trying to answer Packer’s question about how much time we still have to wait before calling the deflation thesis utterly wrong… I was not saying what Watsa can or cannot do…

 

Gio

 

[tombgrt]

There is another reason for low interest rates: excess capital.  Historically capital has been destroyed in wars and with disease and famines.  These have not happened on scale since WWII so with the falling of the communist system and the development of wealth by most countries around the world, you have a large amount of capital generation with little destruction.  You also have a turbocharge of the developed world having a safe haven for wealth created in the developing world.  The last time this happened was between the Napoleonic Wars and WWI.  This was a time of low inflation and deflation.  I agree that there will be real deflation but the question for the CPI hedges is will there be nominal deflation.  If we were on a gold standard or had stable currency regime (like Switzerland), there would be deflation but we are not.  The only time we observed CPI inflation was when commodity priced collapsed but the core chugged along at 1.5% per year.  Some say wait and you will see deflation but we have been waiting since 2008.  My question is how much longer do we have to wait for core CPI inflation or under what conditions will the deflation theory be disproved?

 

Sweden is the only example we have of QE in a deflationary spiral in history that actually worked to stop deflation.  You can quibble with the applicability but I know of no other examples of QEs application in history, do you?

 

Packer 

 

Imo extremely low interests rates are engineered by central banks on behalf of governments around the world, simply because they cannot afford to pay higher interests on their debts. Period.

And this might go on until prices stay very low or else go down. Higher interests rates will exacerbate the debt problem, therefore rendering the final outcome even more unpredictable and dangerous.

 

Watsa has always said that in Japan the printing of a lot of money stopped deflation for 5 years after their stock market and real estate bubble burst… The US probably has printed even more money than Japan did! So I guess we still have to wait at least two or three years… And of course the next serious market correction! If financial assets start to go south, that would be very deflationary imo! After that, if we still have no deflation, I would simply let those CPI contracts expire.

 

I don’t understand how the fact Sweden is the only example available means it should be relevant… But probably it is only me! :)

 

Gio

 

 

But I guess Watsa can point at a single example to make his case? What's the difference?

 

He regularly points at two.  But personally I think there's relatively little use looking at any of these examples except for context.  Every economy and policy is different.  There's evidence that QE has worked in the US and evidence that it is not working in Japan (other people would probably say the opposite).  There's a good theoretical argument that what China needs is a tighter monetary policy, which is counterintuitive.  It's useful to know that QE and its equivalents helped in Sweden (if it did - I don't know anything about that).  But it's also useful to know that it's failed elsewhere and caused hyperinflation in yet more places.  Analysis needs to be a lot more complete than just pointing to one example (no offence to Packer or Prem!).

 

Of course and that is why I made the comment. "This time is different" is actually a generally correct statement, each situation really is different. Good post petec!

[Packer16]

I think one commonality amongst Japan and the US in the 1930s is constipation in the banking/finance system.  From my readings of the 1930s, the banking system was badly broken, you had currency shortages in places around the US.  In Japan, you had still have a system where old debts are not cleared from the system.  This leads to money being trapped in zombie companies with no recirculation to other better uses of capital.  This may be a cause or magnifier of the deflation.  In the US and Europe now we have less constipation which will reduce the chance of a deflationary spiral.

 

As to currency confidence issue, I don't see any alternatives to holding the reserve currencies so although in theory QE may cause issues in practice there may be no way to have those issue come to fruition.

 

Packer 

[petec]

I think one commonality amongst Japan and the US in the 1930s is constipation in the banking/finance system.  From my readings of the 1930s, the banking system was badly broken, you had currency shortages in places around the US.  In Japan, you had still have a system where old debts are not cleared from the system.  This leads to money being trapped in zombie companies with no recirculation to other better uses of capital.  This may be a cause or magnifier of the deflation.  In the US and Europe now we have less constipation which will reduce the chance of a deflationary spiral.

 

As to currency confidence issue, I don't see any alternatives to holding the reserve currencies so although in theory QE may cause issues in practice there may be no way to have those issue come to fruition.

 

Packer

 

I think you're absolutely right re the companies although I actually think ultralow rates also causes constipation - we have a lot of companies here that would be bankrupt at normalised rates and until that happens I wonder how well the economy can normalise. 

 

I hope you're right about reserve currencies!  But I do keep thinking that no reserve currency (apart from gold, arguably, and I am no gold bug) has ever remained reserve currency for long.

[Packer16]

For the past 200 to 300 years the reserve currency has been either the pound or the US dollar so unless you are talking about the real long term, it has been pretty stable.

 

Packer

[petec]

For the past 200 to 300 years the reserve currency has been either the pound or the US dollar so unless you are talking about the real long term, it has been pretty stable.

 

Packer

 

But both currencies were backed by gold until 1971, and since then (it seems to me) we have had a slow loosening of monetary policy.  Confidence is a funny thing: it's there until it's not.

[giofranchi]

For the past 200 to 300 years the reserve currency has been either the pound or the US dollar so unless you are talking about the real long term, it has been pretty stable.

 

Packer

 

But both currencies were backed by gold until 1971, and since then (it seems to me) we have had a slow loosening of monetary policy.  Confidence is a funny thing: it's there until it's not.

 

I don’t know exactly why we are talking about this… but here I agree with Packer: I don’t see any plausible substitute for the USD in the foreseeable future.

 

Gio

 

[petec]

For the past 200 to 300 years the reserve currency has been either the pound or the US dollar so unless you are talking about the real long term, it has been pretty stable.

 

Packer

 

But both currencies were backed by gold until 1971, and since then (it seems to me) we have had a slow loosening of monetary policy.  Confidence is a funny thing: it's there until it's not.

 

I don’t know exactly why we are talking about this… but here I agree with Packer: I don’t see any plausible substitute for the USD in the foreseeable future.

 

Gio

 

I brought it up because of Packer's comment about huge demand for fixed interest securities, which is something that could change overnight (I am not predicting that it will).

 

And I suspect we'd have said the same in the UK until relatively shortly before the pound ceased to be the reserve currency.  Again, not a prediction, but it's always useful to think about unexpected things that could genuinely shake the world.

 

My broader point is that QE may destroy capital and be deflationary.

[watsa_is_a_randian_hero]

The one disconnect I see with FFH thesis is low inflation to slight deflation and stock multiples of low to mid teens.  This would imply and ERP of 7% (very high by historical standards).  If they are right about low inflation lets say 1% (which I agree with), a 1.5% LT bond premium and a more normal ERP of 4 to 5% and a 2% earnings growth rate, it implies a normal stock multiples of 17 to 20 of next years earnings.  We are in that range now.  It is not expensive under FFH's scenario but is expensive for an inflationary environment.

 

The one difference between now and all of the other stock rallies are the low bond yields.  In 2000 and 2007 bond yields were providing reasonable returns versus 0 or negative now. 

 

Packer

 

market multiples is a function of 1/K-G.  K is affected by interest rates and inflation, but so is G.  So inflation shouldn't affect the market multiple to much.

 

I think low interest rates in the current environment are more driven by a low REAL rate of interest.  Real rates of interest on the short end of the curve have been negative for several years now.  I think this is reflective of (1) anti-business/anti-growth political environment and (2) extreme risk aversion on the part of some investors.

[TwoCitiesCapital]

The biggest indicator of deflation for me is simply the velocity of money. It's been dropping every quarter since 2008 and doesn't seem to have stopped. This essentially represents the multiple applied to the entire monetary base in it's effectiveness at stimulating growth, liquidity, and inflation. An expansion of the velocity of money is similar to expansion of credit in this regard.

 

If the velocity of money is still decreasing, it suggests an unhealthy economy and a drag on inflation.

 

I don't know the fancy economic formulas to fool you into believing it, but it's hard for me to get to the inflationary side of things in an American recovery if each dollar is circulated less and less each year.

[barsax]

Good point on the drop in velocity of money.  Pretty surprising PPI  numbers this morning...-.5% PPI/Final Demand month-over-month change.  This number came on the heels of a -.8% drop last month.  Consensus was for an increase of .3%.  Excluding food & energy there was also a decline of -.5% month-over-month.  Consensus was for an increase of .1%... all of that to say this is NOT all down to falling energy prices.  Energy prices are playing a big part, but there's more to this drop than energy alone.