cmlber

I'm at ~18%. I think iOS concerns are overblown and while they may be a short-term headwind, I think long-term it pushes more commerce into Shops which is hugely positive for FB. For 30%+ top line growth it's a very reasonable multiple of earnings; earnings that are burdened by many billions of dollars of expenses that are either (1) deflationary in the long-run (will human content moderators be a thing in 10 years? isn't the server cost to support one user always going down?) or (2) generate no revenue now and someday likely will (VR/AR/messaging). I also think persistent terminal value concerns are way overblown. Instagram / FB will just keep copying successful features in other app like they've been doing.

KC, how do you see the aggregator threat playing out over the next 3-5 years? Just comparing pricing of Papa Johns / Pizza Hut / Marco's on their website vs Doordash, it looks like the customer ends up paying 15-40% more as a result of ordering through Doordash. Even Chipotle who has the as much bargaining power as any brand on Dash you end up paying 10% more ordering delivery through Dash vs direct through Chipotle, on top of the 15% more you pay for delivery from Chipotle direct vs in-store pickup. Seems like eventually once the aggregators consolidate and eliminate discounts the market for 3rd-party delivery will shrink. The pandemic has been a huge boost for them since people are less likely to compare delivery vs pickup/dine-in pricing. Last night I ordered Papa Johns from their website and through Doordash, and Doordash came in 35mins whereas Papa Johns direct came in over an hour. Maybe this is a one off experience, but it seems like giving up control of delivery to Dash was a shortsighted move and that will just mean less delivery density ordering direct from Papa and therefore longer and more variable delivery times which should benefit DPZ over time. They also give up control of the experience. The Dash order got here quicker, but they brought me chicken wings that weren't mine, so somewhere out there someone is pissed they didn't get their wings...

I think Starlink may actually succeed in having a competitive offering to cable, but will be limited by physics in the % of households they can serve. Musk is not known to be conservative in his forecasts and even he says this will only address 3-4% of the world. I think the reason is the bandwidth in a radius is tied to the distance from the satellite which you can’t really change. There’s no way to add more bandwidth to high demand areas (like NYC) since these satellites are evenly spaced out so that there’s continuous service as they orbit. So unlike cable where you put the cables where the houses are and don’t put them where they aren’t, Starlink must have the same shared bandwidth in major cities as it does in Yosemite. You can’t put 4 satellites over NYC and none over Yosemite because (a) they’re always moving and (b) they have to be spaced a certain distance... That’s why it works better as a rural offering.

movys, I see they have tiny market share in the Nordics and are growing rapidly, but it's not because eCommerce is under-penetrated, it's because the marketplace model basically doesn't exist like it does in most other markets. Any idea why that's the case? And why that would just now be changing? Seems like if anything the barriers to shopping at / creating individual websites have come down. Obviously they are growing rapidly, so something is changing, it just seems odd that CDON is the #1 marketplace in the Nordics and has <1% share of eCommerce in 2021...

I think the bigger issue, rather than satellites which might be a threat in Africa but aren't in developed economies, is the ease of putting a tower next to another tower. In the U.S., it's virtually impossible to put a cell tower in close enough proximity to an existing tower to create a true substitute for that radius. I don't know what countries Helios operates in, but I imagine the regulatory barriers don't exist in nearly the same way. Your tenants can credibly threaten to put their own tower 20 feet away, so the pricing power isn't the same.

I think BIPC can be included in indices vs. exclusion of BIP. Given the greater importance of indices this should cause BIPC to trade higher. If there was some type of exchangability should reduce the discount. Packer I think the bigger reason is tax differences. As far as I understand, BIPC pays qualified dividends (24% max tax rate) whereas BIP pays ordinary dividends to US investors (41% max tax rate). However, (a) if you aren't a tax payer, BIP is clearly the better long-term hold, and (b) I think that tax differential would disappear under a Biden administration (w/ a democratic senate). I swapped my BIP for BIPC when the split happened and recently swapped back. If Biden wins and raises capital gains taxes to ordinary rates, I think the spread collapses.

Vertu and Cambria both formed in 2006 by current mgmt. Cambria shares up 420% since then, Vertu shares DOWN 62% since then. That’s all you need to know imo. I don’t trust roll ups that destroy 62% of your capital over a 14 year period to allocate capital well in the future...

Scale is the moat, trust/brand/stickiness all make it difficult for competitors to get to Schwab scale. They are going to eliminate 60-65% of TD's cost base as a merged entity. Said another way, a competitor with $1.3 trillion in assets (not small...) has to operate with $2 billion of expenses that Schwab won't have... The only real scaled competitor is Fidelity and there's no reason for them to compete aggressively to steal each others clients. JPM/BAC/MS all have ~$3T to Schwabs $5.2T (stealing 5-7% market share every year) in client assets, but with business models that are destined to lose clients to Schwab and Fidelity.

We already know what WB did last quarter with regards to his bank holdings. The large holdings have market values as of 3/31 in the 10-Q, from that, it didn’t look like he was a seller. And he filed a late April 13G showing him buying more US Bank.

Doesn't the Fed lending essentially unlimited amounts of money to small & medium sized businesses take mass defaults off the table? Or at least make it MUCH less likely?

Seems like BIP is being thrown out with the BAM bathwater. Sure, BPY has serious issues, but how is BIP yielding almost 7.5% right now? Most of the cash flows come from very long-term contracts with investment grade counter-parties and the debt is non-recourse at the asset level. 1 or 2 of those assets might blow up, but the stock has almost been cut in half even though interest rates have fallen 100bps... Am I missing something? I guess relative to credit in general it's not as attractive given whats happened to non-govt bond prices in a lot of sectors, but seems like very good value right here.

Not sure what #s you're looking at, I have it at about 26x FY20 earnings, not EBITDA. I think business as usual is the catalyst higher: 3-4% price, 3-4% volume, + some margin expansion and layer on leverage and special dividends.

In my opinion, pension liabilities are way worse than debt. Essentially it is debt with a high coupon (the expected rate of return 7-8%) and a maturity of 30 years or so. Debt is easy and cheap now, pension are not. Pension liabilities actually get worse as interest rate go lower. The coupon on it is not 7-8%, it’s 4%, which is lower than GMs longer term bonds yield. And just as the liability gets worse as rates go lower, it gets better as rates go up. In the long run that aspect of it is a wash imo. It’s also somewhat naturally hedged since rates going down means new cars are easier to finance.

I'm trying to see your perspective so let's see if we can turn this into a two-step argument: 1-what happens when realized return is less than expected return (with the return being above or below the discount rate) and 2-what this means for GM. 2- will be discussed if we can agree on 1 :) The discount rate for obligations and the expected return on invested assets are remotely related conceptually but they are two different things and I submit that you are overlooking what is bypassing the income statement. Using the real-life example described above, If I were to buy your education fund obligations, I would use a discount rate in correlation with your overall counterparty risk as a sponsor, not in correlation to your asset allocation related to your investments managed in separate accounts. This is why discount rates for the liability side usually correspond to highly rated bonds. It would be possible to try to match the liability and the asset side with a similar risk profile but it would be difficult and that's not the way pensions are typically funded. As mentioned above, whatever discount rates used for the pension obligation, the difference between realized returns and expected returns, up to a 10% value of assets or liabilities corridor, bypasses the income statement and the net liability (or net asset) number but ends up in equity through OCI. This makes sense because a certain degree of smoothing is reasonable and if, over time, the expected vs realized mismatch balance then the end result is neutral. However, if the company, on a net basis, keeps on building actuarial losses because of realized returns that are lower than expected returns (discount rate for liabilities does not matter here), eventually very true negative cashflows will need to occur to balance the books. This is very insidious because companies can 'manage' this aspect for a very long time and 10% of the greater of the defined benefit obligation or the fair value of the plan assets is a very large number. If interested, I suggest to look at GE's pension management, accounting and reporting. Where am I wrong? I agree if you're valuing GM off EPS, then the expected return matters because as you said it impacts "earnings." Looking strictly at EPS, in part 1 of your question, you and I agree that at some point EPS will be lower in the future to account for "over-earning" in the past with an unrealistic expected return on plan assets. But I don't value GM on EPS but instead value it off FCF. I also account for the net pension liability as something close to, but not quite as "bad" as, debt in adjusting my EV. It's like low quality float, it has a reasonably low cost (the discount rate) but never really matures so long as the business continues to operate and you're well funded to the point of not requiring mandatory contributions. If you start the year 92% funded with an expected return of 7% and only earn the discount rate (plus service cost / benefit obligations to be exact, but that's negligible in this case), you'll still end the year 92% funded. So to the extent a 92% funded plan doesn't require mandatory contributions, it has no cash flow implications to earn less than the expected return so long as you're earning above the discount rate. My main point to begin with though was that the expected return on plan assets isn't used to calculate the net liability on the balance sheet.

The net liability is calculatedly based on market value of assets minus actuarial value of liabilities (which uses something like long term AAA corporate bonds as the discount rate). The 7-8% assumption has no relationship to the current funded status on the balance sheet, it only has implications for future changes in funded status. But as long as they aren’t required to contribute cash because the plans are sufficiently funded, if they earn more than the discount rate (3%ish) on liabilities the funded status should improve over time. You may want to review some of your assumptions here. :) Whether GM's choice of expected return on plan assets of 6.09% or 6.61% is felt appropriate or not given their asset allocation is a separate discussion. Under US GAAP, the future periodic pension cost will be derived using the "expected" return assumption but if the realized return is less than predicted (even if higher than the discount rate used for obligation), this will result in a potentially significant shortfall in the funded status that may not escape the amortization of actuarial deviations as presently allowed. Numbers are large and even small changes in assumptions can have large effects. ... Correct, to quote you, "future periodic pension cost" is derived from the expected return on plan assets, not todays balance sheet liability. Isn't the periodic pension cost equal to service cost (a pretty small number relative to the liability) + interest cost (the discount rate x liability) - expected return on plan assets? To your question: yes but, to be comprehensive, you have to adjust for employer contributions and deal with remeasurements. And the interest cost is based on the net liability (or asset if applicable). Let's use a real-life example. You want to fund an education account for your kids. For the obligation part, you have to take into account the equivalent of the service cost, interest effect, expected tuition inflation etc. In order to establish your yearly contribution, you have to assess your expected return. Let's say you want to get 100K$ in 20 years and you "choose" 7%. Your contribution is 2439$ per year. If your realized return though is 5%, you should have contributed 3024$ per year (+24%, per year). If your realized return is 3%, you should have contributed 3722$ per year (+53%, per year). The remeasurement exercise on the pension asset side ultimately allows to recognize or reconcile the difference between expected and realized returns. GM recognizes the actuarial changes in OCI (numbers are large and potentially very large and it's unclear where they are in the corridor at this point). Back to our real-life example, if you realize along the way that your funding level is too low, a conservative way would be to increase your contributions above minimum levels. Another strategy would be to put 24% of your invested assets into hedge funds. :) I am familiar with GM to some extent and looked rapidly for this post (so this is not audited). Average pre-tax profit in the last five years is about 8,5B. Average periodic pension cost (recognized in P&L) in the last 3 years is about 1,3B. On the surface not life-threatening but significant nonetheless and we are talking about a negative impact that would last a very long time. Interest cost is not based on the net liability, it's pension benefit obligations times the discount rate. So, to the extent your returns on plan assets equal the discount rate, why would the funded status change? Required contributions may change, but the balance sheet funded status would not. Also, GM has recorded pension income, not cost, in recent years; exactly because of the point above, returns on plan assets have been higher than the discount rate. This is one of the larger reasons why "earnings" have been so much higher than auto FCF.