Fairfax 2026

[nwoodman]

4 minutes ago, djokovic1 said:

The biggest thing the market misses is assuming that a soft market means Fairfax compounding will slow down. The 15%+ compounding is not dependent on hard / soft market i.e premium growth. If premium growth slows down (as it has been) the equity will go towards buybacks (as long as the share price is cheap). You win either way.

This is the right way to look at it. They’ve made it clear that Fairfax is a capital allocation machine, with per-share value creation as the primary objective. Float is an important source of leverage, and the fact that its cost has turned significantly negative is a real bonus (and credit to Andy), but it is secondary to where that capital is deployed, and it is far from their only source of capital.

 

Give it time and it will work because, when capital is allocated rationally, the economics are difficult to escape. For a company like Fairfax, particularly in the current environment, short-term share price signalling is a very poor indicator of future returns.

 

I would go a step further and say I am arguably more bullish on the next generation than many others. One of Fairfax’s greatest strengths is that its mistakes have been visible, painful and well documented. The scars from the hedging years, overconfidence in macro calls and periods of excessive complexity are not hidden away, they are worn almost like war medals. In a good sense, they have become institutional knowledge. That leaves me more confident that the next generation will focus on disciplined underwriting, rational capital allocation and per-share value creation rather than trying to prove how clever they are.

 

Anyone who has had the pleasure of cornering Peter Clarke after an exhaustive AGM session, only to find him still willing to engage thoughtfully and answer candid questions, will know what I mean. The same applies to Brian Bradstreet and Kleven Sava, who seem perfectly happy to keep discussing fixed income long after most shareholders have left Roy Thomson Hall. In my experience, that attitude extends across all of the leadership team. It’s a tough gig, insurance is brutal and capital allocation is unforgiving, but they genuinely love their work!

 

What stands out is not that they always have the right answers. It is that they appear genuinely curious, intellectually honest and deeply engaged in the craft of capital allocation. They will make mistakes, every allocator does. Ultimately, each investor has to decide for themselves whether management deserves their trust. I have come to the conclusion that I would rather own a company run by rational people playing a long game than one that simply provides comforting answers in the short term. Everyone’s mileage differs, that’s what makes a market

[Viking]

2 hours ago, djokovic1 said:

The biggest thing the market misses is assuming that a soft market means Fairfax compounding will slow down. The 15%+ compounding is not dependent on hard / soft market i.e premium growth. If premium growth slows down (as it has been) the equity will go towards buybacks (as long as the share price is cheap). You win either way.

I like to use Berkshire Hathaway’s performance in the 1980’s and 1990’s as the best comparable for Fairfax today.

• Not for the exact rate of return that Buffett delivered - although I think Fairfax will be able to compound at a more than acceptable 15% over the next 5 years (as far as my crystal ball looks). 
• But for the fact that he was able to compound at a very high rate of return even in pretty brutal soft insurance markets. 

Why was he able to do this? As others have mentioned - business model and capital allocation is the key. 
 

Another important factor that is new for Fairfax is how they have been optimizing operations:

• Insurance - under Andy
• Equities/wholly owned companies

The result is Fairfax’s operations are now generating much more free cash flow than they have historically. Just look at Eurobank today and compare it to 5 years ago… This is a big deal. I don't think it is well understood. 
 

Greater free cash flow + exceptional capital allocation = strong future returns.
 

The flywheel is spinning very fast right now…

 

Much of the analysis I see on Fairfax is very narrow. It is focussed on a specific factor like  insurance market cycle or interest rates or equity markets or historical factors. It is missing the forest for the trees.

 

It probably gets to Fairfax’s fundamental problem: it is not well understood. If people on this board don’t get it - the larger investment community certainly will not.

 

PS: Fairfax keeps morphing… a lot has been going on under the hood over the past 5 years. Accounting results are understating economic results. To be fair, we all still have lots to learn. 

Edited 8 hours ago by Viking

[Viking]

2 hours ago, Jaygo said:

I would assume the buybacks are absolutely ripping right now. If you look at the buyback kings like AutoZone it was steady buybacks for 25 years.

 

It is conceivable that were under 20 million this year and much lower share count a few years out so if this process has basically just started within the past 5 years we are in for a show. Good things take time.

 

We may bicker about value, ROE, wives and performance but the thought of FFH at 15 million shares outstanding feels pretty good if you look at AZO since 1998. The kicker is that AZO has taken on a lot of debt but I feel Fairfax will handle this better so the result could potentially be more impressive. FFH 10k anyone?

 

 

@Jaygo, I agree... I think Fairfax will continue to be very aggressive on the share buyback front. At current prices/valuation I think a reduction of ~1 million shares per year is likely. 

 

I also think Fairfax will stay undervalued. Perhaps not 1.25xBV undervalued. But undervalued enough that Fairfax will be happy to keep buying back shares in volume. 

 

It certainly will be interesting to see where the share count is in 5 years time. 

[Spekulatius]

10 hours ago, Hamburg Investor said:

 

Historically it came from dividends (spiking after 1929) and inflation (deep deflation). 
 

With dividends reinvested and looking at real returns recovery happened as soon as 1936 (!) instead of 1954. It went down a bit below the recovery line after 1936 until 1944 again; in parts wwii was a major reason.

 

the outcome was over 3x in 1954 with dividends reinvested (real) and over 5 times (nominal) compared to the chart without dividends.

 

Of course that’s only history. And it’s without taxes, costs etc. 

 

 

 

Dividend yield of the SPY ~1%. The dividends won’t bail you out any more.

[Viking]

1 hour ago, nwoodman said:

This is the right way to look at it. They’ve made it clear that Fairfax is a capital allocation machine, with per-share value creation as the primary objective. Float is an important source of leverage, and the fact that its cost has turned significantly negative is a real bonus (and credit to Andy), but it is secondary to where that capital is deployed, and it is far from their only source of capital.

 

Give it time and it will work because, when capital is allocated rationally, the economics are difficult to escape. For a company like Fairfax, particularly in the current environment, short-term share price signalling is a very poor indicator of future returns.

 

I would go a step further and say I am arguably more bullish on the next generation than many others. One of Fairfax’s greatest strengths is that its mistakes have been visible, painful and well documented. The scars from the hedging years, overconfidence in macro calls and periods of excessive complexity are not hidden away, they are worn almost like war medals. In a good sense, they have become institutional knowledge. That leaves me more confident that the next generation will focus on disciplined underwriting, rational capital allocation and per-share value creation rather than trying to prove how clever they are.

 

Anyone who has had the pleasure of cornering Peter Clarke after an exhaustive AGM session, only to find him still willing to engage thoughtfully and answer candid questions, will know what I mean. The same applies to Brian Bradstreet and Kleven Sava, who seem perfectly happy to keep discussing fixed income long after most shareholders have left Roy Thomson Hall. In my experience, that attitude extends across all of the leadership team. It’s a tough gig, insurance is brutal and capital allocation is unforgiving, but they genuinely love their work!

 

What stands out is not that they always have the right answers. It is that they appear genuinely curious, intellectually honest and deeply engaged in the craft of capital allocation. They will make mistakes, every allocator does. Ultimately, each investor has to decide for themselves whether management deserves their trust. I have come to the conclusion that I would rather own a company run by rational people playing a long game than one that simply provides comforting answers in the short term. Everyone’s mileage differs, that’s what makes a market

 

+1

[Spekulatius]

1 hour ago, Viking said:

I like to use Berkshire Hathaway’s performance in the 1980’s and 1990’s as the best comparable for Fairfax today.

• Not for the exact rate of return that Buffett delivered - although I think Fairfax will be able to compound at a more than acceptable 15% over the next 5 years (as far as my crystal ball looks). 
• But for the fact that he was able to compound at a very high rate of return even in pretty brutal soft insurance markets. 

Why was he able to do this? As others have mentioned - business model and capital allocation is the key. 
 

Another important factor that is new for Fairfax is how they have been optimizing operations:

• Insurance - under Andy
• Equities/wholly owned companies

The result is Fairfax’s operations are now generating much more free cash flow than they have historically. Just look at Eurobank today and compare it to 5 years ago… This is a big deal. I don't think it is well understood. 
 

Greater free cash flow + exceptional capital allocation = strong future returns.
 

The flywheel is spinning very fast right now…

 

Much of the analysis I see on Fairfax is very narrow. It is focussed on a specific factor like  insurance market cycle or interest rates or equity markets or historical factors. It is missing the forest for the trees.

 

It probably gets to Fairfax’s fundamental problem: it is not well understood. If people on this board don’t get it - the larger investment community certainly will not.

 

PS: Fairfax keeps morphing… a lot has been going on under the hood over the past 5 years. Accounting results are understating economic results. To be fair, we all still have lots to learn. 

In soft market, earnings will go down. I did a rough calculation and at a 100% combined ratio (no underwriting profit), FFH would still be doing ~10% ROE and maybe a shade better. If the premium growth is zero, they could be buying back 10% of the shares theoretically without depleting their capital base. Their historic combined ratio is 95% through the cycles so the base case would be higher than 10% ROE.

 

I think my forecast 10% is a quite pessimistic take and they should be able to do better than  this even in a appt insurance market. This assumes that interest income and equity income does not vaporize.

 

The only way this is going badly if their underwriting start to suck and their investment go sour, due to a prolonged bear market.

[Maverick47]

2 hours ago, Viking said:

I like to use Berkshire Hathaway’s performance in the 1980’s and 1990’s as the best comparable for Fairfax today.

• Not for the exact rate of return that Buffett delivered - although I think Fairfax will be able to compound at a more than acceptable 15% over the next 5 years (as far as my crystal ball looks). 
• But for the fact that he was able to compound at a very high rate of return even in pretty brutal soft insurance markets. 

Why was he able to do this? As others have mentioned - business model and capital allocation is the key. 
 

Another important factor that is new for Fairfax is how they have been optimizing operations:

• Insurance - under Andy
• Equities/wholly owned companies

The result is Fairfax’s operations are now generating much more free cash flow than they have historically. Just look at Eurobank today and compare it to 5 years ago… This is a big deal. I don't think it is well understood. 
 

Greater free cash flow + exceptional capital allocation = strong future returns.
 

The flywheel is spinning very fast right now…

 

Much of the analysis I see on Fairfax is very narrow. It is focussed on a specific factor like  insurance market cycle or interest rates or equity markets or historical factors. It is missing the forest for the trees.

 

It probably gets to Fairfax’s fundamental problem: it is not well understood. If people on this board don’t get it - the larger investment community certainly will not.

 

PS: Fairfax keeps morphing… a lot has been going on under the hood over the past 5 years. Accounting results are understating economic results. To be fair, we all still have lots to learn. 

Edited 1 hour ago by Viking

Spot on @Viking.  
 

I got my start in the insurance business in 1990.  For the entirety of my career, it was a mistake to confuse Berkshire with any competitors of the insurance company I worked for.  In fact, quite often metrics for the US or global insurance industry had to be adjusted to exclude Berkshire lest they distort things and give too rosy a picture of how well the industry was doing.  Berkshire simply had much more capital supporting its insurance premium writings than any other insurance company because of its wholly owned subsidiaries and the outsized growth of its investments in equities.

 

A typical pure play insurance company will “lever” its shareholder equity/policyholder surplus such that each dollar of it supports the writing of $2 or even $3 of premium.

 

Every dollar of Berkshire equity nowadays supports much less than a dollar of premium volume.

 

If we look at Fairfax today its premium volume is about $33 billion, and its shareholder equity is probably just a few billion shy of that…but as Prem reminded us, there is almost a $4 billion of fair market value to carrying value cushion that doesn’t yet appear on the balance sheet.

 

So the Fairfax ratio is probably a 1 to 1 now…which makes comparisons with other pure play insurance companies much less appropriate.

 

Fairfax is indeed metamorphosing before our eyes.  It’s already a mistake to consider it only an insurance company, and ten or twenty years from now we are likely to consider it more as a conglomerate with a strong insurance business, even more immune to the vicissitudes of the insurance cycle than companies that are solely insurers than it already is today.

Berkshire has normalized operating earnings of over $40 billion a year now.  Who cares if they might occasionally wipe out a quarter’s worth of earnings in the event of their $10 billion share of a massive $300 billion industry catastrophe?
 

Likewise, with $5+ billion of normalized annual operating earnings for Fairfax, who cares if they might experience a $1 or $2 billion catastrophe event?  
 

Large future catastrophes that could be balance sheet events for the typical insurance company are likely to be merely quarterly earnings/income statement events for companies like Berkshire and the new Fairfax.

Folks who are even high level execs in the insurance industry simply don’t understand Fairfax well enough to understand that insurance cycle observations or rules of thumb about how well the industry is likely to perform in a soft market are not as applicable to Fairfax as they are to other insurers.  Nor does Morningstar yet appear to understand that viewing Fairfax solely through the lens of an insurance company investment is increasingly less helpful to consumers of their research.

Edited 6 hours ago by Maverick47

[Viking]

1 hour ago, Spekulatius said:

In soft market, earnings will go down. I did a rough calculation and at a 100% combined ratio (no underwriting profit), FFH would still be doing ~10% ROE and maybe a shade better. If the premium growth is zero, they could be buying back 10% of the shares theoretically without depleting their capital base. Their historic combined ratio is 95% through the cycles so the base case would be higher than 10% ROE.

 

I think my forecast 10% is a quite pessimistic take and they should be able to do better than  this even in a appt insurance market. This assumes that interest income and equity income does not vaporize.

 

The only way this is going badly if their underwriting start to suck and their investment go sour, due to a prolonged bear market.

 

I think Fairfax's P/C insurance business has changed dramatically over the past 15 years:

• Andy has had time to execute his vision
• Expanded aggressively through acquisitions from 2015 to 2017
• Pivoted in India (from ICICI Lombard to Digit)
• Dramatically reduced the size of runoff
• Seeded interesting opportunities like Ki and Digit Life and Re-insurance 
• Expanded aggressively in hard market
• Completed a range of 'bolt on acquisitions': Singapore Re, GIG and Albingia
• Began taking out minority partners: Brit and Allied World
• Reduced its catastrophe exposure (historically speaking)
• Still have the opportunity to take out minority partners in Allied World and Odyssey at very favourable prices

The net result is its insurance business in 2026 is a completely different animal than its insurance business in 2011. It is remarkable all the different things Fairfax has accomplished over the past 15 years (there is a very important lesson here... Fairfax was very aggressive in growing and improving the business in a soft market...).

 

This makes it very difficult to use historical numbers for insurance for Fairfax that are pre-2011 (CR etc). So let's look at the past 15 years. How many years did Fairfax have a CR over 100? I think it was one. And it was caused, not surprisingly, by historically large cat losses. Most of the years from 2011 to 2026, insurance was in a soft market. Hard markets are the exception.

 

Fast forward to the next 5 years. Will Fairfax write at an average CR of 100? 

 

Absent a historic level of catastrophes over multiple years, I don't see it. Not as a baseline forecast.

 

Importantly, there is a high likelihood that reserve releases will trend higher than normal in the coming years (on average - with some volatility by quarter). That is usually what happens when we come out of a hard market. This will be a tailwind to the CR.

 

Again, I just don't see how we get an average CR of 100 over the next 5 years. Building that into a baseline forecast isn't being conservative - it looks extremely bearish. And I am not sure how that approach ever leads to optimal results for an investor.

Edited 5 hours ago by Viking

[SafetyinNumbers]

1 hour ago, Spekulatius said:

In soft market, earnings will go down. I did a rough calculation and at a 100% combined ratio (no underwriting profit), FFH would still be doing ~10% ROE and maybe a shade better. If the premium growth is zero, they could be buying back 10% of the shares theoretically without depleting their capital base. Their historic combined ratio is 95% through the cycles so the base case would be higher than 10% ROE.

 

I think my forecast 10% is a quite pessimistic take and they should be able to do better than  this even in a appt insurance market. This assumes that interest income and equity income does not vaporize.

 

The only way this is going badly if their underwriting start to suck and their investment go sour, due to a prolonged bear market.

What over/under on the combined ratio average for the next 5 years would you be willing to take the under for even odds?

 

I think there are some mitigating factors that investors are ignoring:

 

1. Reserves are stacked on the casualty side and those should be released over the next 5 years

2. The business should have seen operating leverage from the premium growth and based on disclosure a lot of that is being spent on technology. Presumably that has some payback on productivity but it could just be table stakes.

3. Fairfax is more diversified than its peers. It can lean into its international operations which don’t operate on the same insurance cycle.

 

[HoldForDearLife]

15 hours ago, SafetyinNumbers said:

What’s the build up of the 10% (say 13.3% pre-tax ROE)? Given the coupon on the fixed income portfolio is 5%, it gets us to about a 10% pre-tax ROE. Assuming underwriting covered head office and leverage expenses, the equity portfolio would have to return 3.3% to meet the hurdle. That seems like a low bar given the state of fair value over carrying value and the high earnings yield of equity accounted for and consolidated positions. 
 

Tney also have the call options on Allied World and Odyssey minority interests which are very accretive. 

That's underwriting a 95% CR, no? Not egregious, but that's not exactly a "nothing" assumption either in a softening market. If you were to underwrite at a 100% CR instead, you'd need the return on the equity portfolio to be around 8.0 % to offset the drop in underwriting income. 

 

That said, one of the many reasons I like Fairfax is that you're in no way dependent on terrific underwriting for the investment to make sense. If we go five years at a low 100s combined ratio, we're probably going to remain profitable and suffer less than competitors. We have the cash to survive a bad year with catastrophes. Compared to many other insurers with combined ratios in the low 80s, I do very much prefer "our expectations" over them. 

[Hamburg Investor]

22 hours ago, Spekulatius said:

Yes,  but where does the money come from to average down in the Great Depression?

 

5 hours ago, Spekulatius said:

Dividend yield of the SPY ~1%. The dividends won’t bail you out any more.

 

I was just referring to the past.

 

Yes, averaging down is harder today; on the other hand buybacks were around Zero in the 1930s if I am right. And today it’s at 2.2%. And buybacks are not so different from averaging down - you’re getting a bigger part of the whole (and it’s even more tax efficient for A lot of investors than divs).

 

So together that’s more like 3.2%. That can easily grow to 4% or more in a crash scenario. Not great, but if you find good businesses at reasonable prices (I‘d happily buy FFH 20% off from Fridays close…  ) not bad either. 

 

If one is a longterm holder of stocks and doesn’t have to sell for 5 to 10 years, a normal bear market will not end in a disaster imho. Time and Buybacks (and a little bit of divs) will help to head even stronger into the next bull market. If you have to sell bigger parts of your portfolio within 5 years - that’s not so good. 

[djokovic1]

7 hours ago, SafetyinNumbers said:

Mr. Market Structure doesn’t care about returns, he cares about revenue growth. Then investors who do care about returns but think the market is efficient use analogs to predict forward returns. Of course, they could be right, but the odds seem really low which is especially true when the timeline is extended. 

 

My point is different. Its that regardless of what multiple the market assigns Fairfax in the short term, investors will get 15%+ compounding of EPS in both hard and soft markets (with bumps both up and down along the way).

 

If one understands that, then it also follows that Fairfax is a no-brainer investment at 8x EPS today. Separately, I do also think the multiple is much more likely to go up than down given the starting point today and as the evidence of compounding keeps stacking up. Long term weighing machine, short term voting machine etc...You just don't know when.

[djokovic1]

7 hours ago, nwoodman said:

This is the right way to look at it. They’ve made it clear that Fairfax is a capital allocation machine, with per-share value creation as the primary objective. Float is an important source of leverage, and the fact that its cost has turned significantly negative is a real bonus (and credit to Andy), but it is secondary to where that capital is deployed, and it is far from their only source of capital.

 

Give it time and it will work because, when capital is allocated rationally, the economics are difficult to escape. For a company like Fairfax, particularly in the current environment, short-term share price signalling is a very poor indicator of future returns.

 

I would go a step further and say I am arguably more bullish on the next generation than many others. One of Fairfax’s greatest strengths is that its mistakes have been visible, painful and well documented. The scars from the hedging years, overconfidence in macro calls and periods of excessive complexity are not hidden away, they are worn almost like war medals. In a good sense, they have become institutional knowledge. That leaves me more confident that the next generation will focus on disciplined underwriting, rational capital allocation and per-share value creation rather than trying to prove how clever they are.

 

Anyone who has had the pleasure of cornering Peter Clarke after an exhaustive AGM session, only to find him still willing to engage thoughtfully and answer candid questions, will know what I mean. The same applies to Brian Bradstreet and Kleven Sava, who seem perfectly happy to keep discussing fixed income long after most shareholders have left Roy Thomson Hall. In my experience, that attitude extends across all of the leadership team. It’s a tough gig, insurance is brutal and capital allocation is unforgiving, but they genuinely love their work!

 

What stands out is not that they always have the right answers. It is that they appear genuinely curious, intellectually honest and deeply engaged in the craft of capital allocation. They will make mistakes, every allocator does. Ultimately, each investor has to decide for themselves whether management deserves their trust. I have come to the conclusion that I would rather own a company run by rational people playing a long game than one that simply provides comforting answers in the short term. Everyone’s mileage differs, that’s what makes a market

 

Great post and completely agree. Also great to meet you and @Viking in Toronto. The highlight of my trip was the shareholder dinner Q&A with Prem and team, it was special. 

[Hamburg Investor]

5 hours ago, Maverick47 said:

A typical pure play insurance company will “lever” its shareholder equity/policyholder surplus such that each dollar of it supports the writing of $2 or even $3 of premium

Isn‘t float the lever, not premiums? And float is more in the direction of double equity (less, if we count in all the hifden value). And 2 to 1 is relatively high leverage, isn’t it?

 

5 hours ago, Maverick47 said:

Fairfax is indeed metamorphosing before our eyes.  It’s already a mistake to consider it only an insurance company, and ten or twenty years from now we are likely to consider it more as a conglomerate with a strong insurance business, even more immune to the vicissitudes of the insurance cycle than companies that are solely insurers than it already is today.

Yes, I guess the float leverage should go down over the years. But there are some ingredients in FFH, that are quite different to BRK:

 

1. FFH: A real worldwide insurance foot print

I am pretty sure, that FFH has the widest geographical insurance business. They are on all continents. FFHs premiums come not  lnly from the USA an Canada; they come from Ukraine, Arabia, Poland, Asia, South Africa, Botswana, Vietnam, Singapore, Brasil, Colombia, Uruguay, UK, Germany, France, Hongkong, Indonesia, Sri Lanka, … It‘s over 30 countries. This might be a great opportunity to keep growing, as there might be opportunities in strong growth countries. Like India, but there are a lot of other emerging markets in Asia and orher parts of the world. BRK never focused its insurance businesses in other countries - I guess it’s below 10. At FFH going to a lot of countries is a strategy. It’s great to be, where the world grows. I am pretty sure, FFH will buy more Insurance businesses and grow that number actively over the years. GIG was not the last deal to grow thhe footprint.

 

2. FFH: Smaller insurance businesses than BRK

FFH has 20% of its insurance businesses in mostly very small businesses. These are - again - the ones located in other countries. But the argument for growth here is not the same: Small businesses tend to grow stronger than big ones. Yes, big portions of the premiums are in relatively big insurance companies like Zenith, Allied, Crum & Forster, Odyssey, Northbridge. But still 20% is in those small insurance businesses - and we see this portion growing from year to year. Be it alone - or through investments from FFH itself.

 

 

2. Culture & Learning from others.

Apart from being located in 30+ countries, which are partly growibg strong, and apart from buying in into new countries lver the years, and apart from the fact, that those small businesses tend to grow stronger than the big ones, culture and the opportunity to leanr from each other is another growth driver. 
 

Say, you are a small insurance business in a small country. Isn’t it a great opportunity to learn about AI strategies from other insurance subs these days and maybe even share cost for developing new tools? An insurer in Botswana clearly is no competitor for an indian or Brasil or German insurer. I am pretty sure, that leveraging opportunities like these will help small insurers a lot. I don’t know if that happens today, but why shouldn’t they do it?

 

I guess being an insurer in Botswana, being backed by FFH gives a lot of tailwinds to you and helps keeping the course, handle critical times, get cash to invest, if opportunities arise, keep costs down - and grow even stronger.

 

Imho although I agree, that FFH in 10 and 20 years might be leveraged less, I don’t see the float leverage to come down at the same pace than at BRK.

 

[SafetyinNumbers]

1 hour ago, djokovic1 said:

 

My point is different. Its that regardless of what multiple the market assigns Fairfax in the short term, investors will get 15%+ compounding of EPS in both hard and soft markets (with bumps both up and down along the way).

 

If one understands that, then it also follows that Fairfax is a no-brainer investment at 8x EPS today. Separately, I do also think the multiple is much more likely to go up than down given the starting point today and as the evidence of compounding keeps stacking up. Long term weighing machine, short term voting machine etc...You just don't know when.

I was adding to your point. Explaining why the market doesn’t reflect what you laid out. I think when is the next hard market but I’m not sure anyone else thinks that.