Blessed by the Potato

Meat is an interesting and contentious meal choice. I’m personally a vegetarian, but a lacto-ovo one. I don’t have any particularly strong moral objections to using animals for human purposes: though I personally would prefer not to get involved in the process, I don’t throw a fit every time a friend has a BBQ (and my daughter is an omnivore until she decides otherwise). Yet I think people eat vastly too much meat. So many animals raised just for food, it’s so wasteful. Grazing animals can better make use of marginal farm/pastureland/scrub, but many acres of prime farmland is also given over to raising animals or growing feed for them. A laying hen can produce hundreds or thousands of eggs in its lifetime, but you consume way more chicken than that ratio of eggs:meat would suggest should be in your diet (generally you don’t eat the old egg-layers, you grow chickens just for meat and a separate group for eggs that you then feed to your dog).

For cows, it takes 5-100 times as much inputs like land, water, fossil fuels, and grains to make a pound of beef as it does to make a pound of vegetarian feed (in some reports these figures are adjusted for the protein content, so a report that says that it takes 10 pounds of vegetable protein to make 1 pound of beef protein may mean that ~100 pounds of feed went into the pound of meat). The rough rule of thumb I have heard is that we can feed about 10 times as many people on a vegetarian diet as a heavily meat-based one given the same input constraints. Whatever the exact number is, it’s a lot: there’s a fair bit of waste putting a cow between you and your food. I like to cut out the middle man.

The announcement that lab grown meat has been made (and eaten) made the news recently, spurring a lot of speculation about what the future will bring. I think it will be interesting, particularly for those simultaneously interested in animal cruelty and eating hamburgers, but I don’t think this is ushering a new paradigm of how we feed ourselves, and it’s not an immediately apparent investment opportunity.

The big question in my mind is what will the inputs be under the lab-grown scenario? Is there a business case there?

Land: One of the big advantages of lab-grown meat is that you don’t need to waste all that space letting the animal move around. You can pack the meat in close and even stack the building up with multiple stories, putting a lot of production into a small footprint. However, land is not expensive. It perhaps is not right, but anyone can just go and buy up farmland — prime farmland — and plop a condo on it, and most of the cost will still come from the construction process. So saving land is great, but we simply don’t value it enough right now to have this become a factor. Indeed, the need for infrastructure will likely make this way of producing a hamburger much more costly than just letting herds of cattle roam across the foothills of Alberta, and the capital investment required will likely keep many from making the plunge out of curiosity.

Water: I honestly don’t know where all the water is used in cattle farming, the reports suggest ridiculous amounts of it in use. Are the cows being bathed nightly? Anyway, water use will likely decline, but unless these meat factories get built in the boonies the water expense will likely remain unchanged as expensive, treated city water gets used.

Feed: Here is my big question. By not growing a full cow (just the meaty parts), and by just growing it long enough to harvest (rather than burning up resources in ongoing metabolism as the cow wanders around the fields day after day for months and years on end), there should be a big gain in the ratio of feed inputs to meat outputs. But it’s still not going to be very close to 1:1 — there are still inefficiencies in growing muscle tissue. There might be a decent efficiency between the growth medium energy and the meat-like substance protein content, but the question is whether that gain helps us any when you consider the full chain. With a full cow you just shovel a bucketful of corn/hay/wheat/oats into a trough and let the cow do its job. With a lab-grown collection of cells you will have to predigest the grain inputs through some kind of process to produce a sterile nutrient solution. What losses and inefficiencies will be associated with that value chain?

I doubt that having a disembodied bovine stem cell grow a muscle cell in a vat is going to come anywhere near the efficiency of eating some corn and growing a muscle cell yourself. Perhaps the ratio will come down, from say 100:1 to 10:1, but given the other costs I don’t think that makes a business case. I suspect that, at the end of the day, there will be next to no advantage here: 100 pounds of unprocessed grains may be cheaper and less resource intensive than distilling 10 pounds of grains down to soluble carbohydrates and an osmotically balanced protein slurry.

Other inputs: What antibiotics and growth hormones will be needed to keep these cells alive and stimulate them into growing? There won’t be as many fossil fuels needed to drive trucks and manure-spreaders, but what about the electricity needs of the incubators and buildings? What of the plastics and glassware?

Manpower: How educated will the meat rustlers of the future have to be? How many cowboy technicians will we need per pound of beef, versus farmhands per pound the old way, and what wages and benefits will they demand?

I simply have trouble seeing this as the future of food.

Food is cheap. Fundamentally it is the most basic and essential need of every person — even the poorest — and we go through a lot of it. I don’t see how lab-grown meat will come close to the price of traditional meat. Even with scaling up and industrializing the process in a factory, will this be competitive? There might be a niche market for PETA members to get their guilt-free burger fix, no matter the price, but will it truly become an industry?

In some dystopian future where process efficiency trumps capital requirements, maybe. But if we’re in a world that’s down to counting every last grain of wheat, why not just eat the plant matter ourselves?

Now, lab-grown organs are not so different from lab-grown meat (indeed, just a specialized form of lab-grown meat). But people will be willing to pay vastly more per pound of life-saving kidney than they would for mediocre appetite-satisfying hamburger. So while I don’t think lab-grown meat will be a food source to count on soon, lab-grown organs might be here sooner (despite the added complexity). Indeed, if lab-grown meat is to become a viable business, it will likely depend on innovations and technologies developed for organ regeneration.

Jenn recently linked to an interesting article about post-doc pay, and how the low pay (and other issues, like the constant moving and uncertainty and short-term contracts and lack of benefits) right at the point where women’s fertility starts to drop is one factor keeping them out of science. Go and read that article, but I think this goes well beyond just women in science, post-docs and starting families.

I keep thinking of ways to dramatically reshape the way we do science. They may not be practical, but I like thinking outside the box from time to time.

One set of related ideas I keep coming back to are the issues of compensation and focus. Grad students and post-docs are paid terribly. How terrible? Well, in my department grad students made about $14k-16k as a base stipend (and that level has not changed in almost two decades, inflation be damned), top students with national scholarships could take home about $33k. Yes, per year, with restrictions on seeking outside work. This is in part because they are said to be trainees who are learning how to be proper scientists. Except if they make it through the funnel and up the pyramid, or whatever visual metaphor you may choose, they teach and write grants and supervise — skills they are largely not being taught.

So the idea I toss around is that of a permanent post-doc, or professional bench scientist: a position for someone who will spend their life doing hands-on research, and who gets paid a professional salary for it.

Along with that would be wage/stipend increases for grad students: there is a lot of catching up to do just to get back to the inflation-adjusted level of poverty they were at a decade ago, let alone getting to the point where it is recognized that they are the driving force behind science, and that a senior PhD student is a professional with years of training and specialized expertise making less than minimum wage. One related option might be to shorten PhD programs — it runs the risk of devaluing the degree, but did the 4th and 5th years of my own slog through grad school add much to my development as a scientist that the 2nd and 3rd years did not already? How has the average time to graduation changed over the past couple of decades?

It’s a tough issue, and would represent massive disruptive changes, with no real advocate to push for it. I’m really not even sure myself if these wild speculations I sometimes have are worth any further consideration at all. I mean, even if that is a place we wanted to move to, how would we possibly get there?

In a sense, science is powered by slave labour. If we restricted entry into grad school so that a higher percentage of PhDs could stay in academia (and let the industries that end up hiring PhDs instead hire MSc grads or some newly-created in-between research-intensive 3-4 year expert degree); or reduced the graduation hurdle so that they only did 2 experiments instead of 3, and graduated before 31 years of age — or really any change along those lines — we would limit the amount of science that could get done on current budgets. Unless we truly were able to hire more efficient and productive talent (or focus and dedicate the talent we have) with the increased compensation, the fact is that less research would get done for today’s research budget. This seems an insurmountable problem.

Then I thought, what if instead of thinking of slavery as a harsh verbal rhetoric, I looked at it as an actual model? After all, that problem has been solved. Slavery doesn’t exist in the modern civilized world, but did at some point in our past. Many countries weaned themselves off, with the US having a particularly dramatic and definite end to the practice after the Civil War. How did the transition work out then? What lessons can we learn for transitioning the economic model of science? Unfortunately I’m not enough of a historian to say, so I will have to end here as some food for thought.

Here is a recent TED talk on philanthropy and advertising you should go watch.

This is an interesting perspective. As someone who is currently “overhead” I can, to a certain extent, agree. Besides my current day job, I also recently picked up an interesting freelance gig. A donor hired me to rewrite and revamp a fundraising brochure for a local hospital. Because I’m being paid directly by the philanthropist and not the foundation, my fees will not appear in their books as overhead, though I hope that the work that I’ve done indeed helps multiply the donations they eventually receive regardless. (You can see a PDF of the brochure here.)[Update: I did a second related one.]

To some extent there is a need for scale in philanthropy. A charity attempting to say fund research looking for a cure for cancer is not going to be able to make much of a dent with an annual budget of $100k — that’s barely one research grant (and even then the lab has to have some other source of core funding). It takes millions to be able to have enough to get together a panel of peer reviewers to examine grant proposals, or to buy expensive pieces of infrastructure such as PET scanners. And that takes some kind of investment to scale up — whether resources for advertising, or the volunteer effort to go viral on the internet.

But there’s a limit. At some point you could just be raising money to pay people to try to raise more money. In the talk he mentions that charitable giving has been stuck at 2% of GDP for decades. If there is some sort of mechanistic reason for that — it’s the amount people are capable of giving, or some sort of unconscious philanthropy budget in the population as a whole — then pushing for more overhead is just shifting the charity spending around, and in fact a net negative due to the overhead. It is possible that, by being able to tackle large challenges smaller organizations could not, we would be better off with one (or a few) massive billion-dollar charities spending a total of $240B than with a bunch of smaller million-dollar charities spending $275B with lower overhead costs. But if outcomes are directly related to dollars spent, more overhead would indeed simply mean more waste.

Consider a parallel with investing: you could pay a brilliant manager some percentage of your funds under management, and they might be able to beat the market for you. But there’s only so much return out there to be had: if everyone else hires an investment manager then everyone is on an even footing and is back to getting basically average returns… less the overhead to the managers.

There were three other points of his I want to discuss.

The first was on compensation. The big unanswered question for me was whether you would get value for that extra $300k spent on talent in his hypothetical. Perhaps everyone is better off if the MBAs pursue for-profit $400k salaries and donate $100k to the charity, who can then hire an $87k/year executive. If the charity tried to hire someone for $400k, would they get more than the ~$300k difference back in value? Charities, after all, don’t have all the things to manage that for-profit businesses do: maybe the extra money buys you advertising and capital markets experience, which you just don’t need as a non-profit. And why doesn’t that logic apply all down the chain? We pay grad students and post-docs a disgraceful pittance for trying to find the cures to our modern medical ailments, but brilliant technically-minded and driven people can make far more in the private sector. Would we have long since solved this pesky cancer problem if we were only willing to retain top talent in the research enterprise by setting post-doc starting salaries at $400k, and grad student stipends at $75k?

The second was on the whole comparison of the for-profit and not-for-profit sectors. You see, the two are very different fundamentally. When I give my money to Coca-Cola for a beverage, or to Amazon for a book, I am transacting with them for something. I don’t care how much they spend on overhead, because I am making my decision on whether or not to give them money based on what they are giving me in return at that moment. I need to only extend a small amount of trust to them (trust that Coca-Cola hasn’t diluted my Coke Zero, trust that my book from Amazon will arrive undamaged in a under a week), and I have recourse if my trust is violated: I can demand my money back, sue for breach of contract, etc. But once I send my money to them and receive my item, it is no longer my money. It’s their money, they can do with it as they wish.

Giving money to a charity is a completely different thing. I’m not getting a thing or a service, I’m giving my money to the charity to make the world a better place. I am trusting them to put my money to good use. Though the money is out of my hands and I have no recourse to get it back once I give, at no point do I consider it “their money” to do with as they please. Maybe they could give me a better “product” if they spent three times as much on overhead as I had reasonably expected — but that is a lot of trust for me to give. While only people donating staggering amounts of money expect to be able to direct their donations precisely, I still expect that, in general, my donation will be used for the stated purpose — ultimately mostly directed towards some kind of program spending rather than churning overhead or as risk capital. Their use and governance of the money will continue to be the concern of the donors, and so there is a very real reason for spending on overhead and risky activities to be perceived differently than in the for-profit sector.

In an analogy to investing, let’s say that there was a company that raised $100M in a stock offering to pursue a business idea. They went out the first few years and spent $90M of the money doing what had to be done (hiring people, renting office space, advertising etc.). After a few years of losing money they discover that the business model is just not viable. To continue is to throw good money after bad so they wind up operations. As a shareholder, through the first few years you would have been obliged to let management take the risk and pursue the business, spending your capital on whatever “overhead” was needed to do so. After it failed, you would expect that any residual money ($10M in this example) would be returned to you as the business was shuttered, and promptly. If they dragged their feet in the wind-up, paying salaries for years, burning through your capital with no purpose you would rightly be pissed at that loss.

In the not-for-profit sector, overhead spending that is going to have a multiplicative effect is difficult to discern from the telemarketer full employment program. How do you know whether you’re in the phase of risk capital spending that is pursuing the innovative business model with lots of potential, versus the phase that is basically the insiders stealing from the other contributors of capital? The risk-reward equation is not the same in the not-for-profit and for-profit examples, and the governance is different: profits are much easier to measure than “impact” or “do-goodery”. And as unethical and despicable as it was for the executives in the hypothetical example to burn the remaining shareholder capital after it was clear nothing would come from it, it is even more morally repugnant to live large off people’s charitable donations — hence the aversion to overhead spending.

And the last point I wanted to discuss was that of spending more overhead as a percentage to scale up. In the presentation he just kind of implicitly assumes that to scale up an organization might have to spend a larger percentage on overhead. But should this be so? Shouldn’t scaling up offer economies of scale? If instead of spending $100 to raise $1000 at a bake sale, a charity should spend $100M on organizers for a massive event and TV air time, then shouldn’t that investment be expected to pull in $1B for program spending, rather than just $250M as his 40% figure would indicate? Now again, maybe the efficiencies come on the spending side (perhaps spending $250M in one organized way with a unifying strategy does more cumulative good than spending $1B in separate $1M chunks).

So while I can see some of his points about needing large-scale charities to tackle large-scale problems, and that sometimes investments have to be made (to train people, to build infrastructure, etc.) and sometimes more overhead has to be spent (for strategy, for advertising), I do not fully agree. Sometimes overhead is just money not going to program spending, and I would hope that scaling up would bring about more efficiency rather than less. The not-for-profit and for-profit sectors are have larger fundamental differences than he suggests. And when you come right down to it, I want to be able to know that my charitable donation is going towards the stated purpose and not to the Canada Foundation for Telemarketer Employment. Maybe looking at the percentage of spending on overhead is not the best way to choose where to donate — perhaps we need some impossible measure of impact per dollar donated — but we will naturally gravitate towards metrics that are easily enumerated.

Well, scam is way too harsh, but the hyperbole may be needed to counteract all the “OMG, Shoppers Optimum is the best rewards system!” that’s out there.

I used to love Shoppers; though they’ve long been a terrible pharmacy (high dispensing fees, and we’ve experienced many frustrations with for example filling prescriptions weirdly to generate even more filling fees), I used to think that they were a pretty terrific all-night convenience store: well-lit, well-stocked, with competitive prices. And hey, Optimum.

Then that all started going away: they jacked their prices through the roof, and stopped being well-stocked on sale items. Optimum points were devalued. Customer service took a hit, and many coupons, bonus points, and sale prices were not going through at the register. The one by our house sells out of flyer items sometime before noon on the first day of the sale. I’ve almost entirely stopped shopping there (even though it’s between the bus and my house, so super convenient). To say that it’s one of the best (or to take out the qualifier entirely) because it has the highest percentage return of any loyalty program misses the catch: it’s only good at over-priced Shoppers.

There have been a few times in the recent past when Wayfare has been induced to spend $50-$100 at Shoppers due to a bonus Optimum points event. As Wayfare puts it, she only shops there on 20X the points events, and redeems her points on bonus redemption days. “It’s like 30% off!” On the surface, that looks approximately correct: you normally earn 10 points per dollar spent (truncated — so $17.40 pre-tax gets you 170 points), and points are “worth” $0.00179 each if you save up to the top tier. That works out to about 1.79% return on normal days, and as much as 42% if you only buy on 20X the points days and only redeem on bonus redemption days, and then spend exactly $200 (and not pick up a bunch of overpriced crap that you pay for at the regular rate once you’re past $200). Compare that to the ~1% you get with PC points (and only then if you qualify for the special card) or the 0.5% from Air Miles (though it may be even less than that now with the constant devaluing of points).

Here’s my thinking: yes, you get “money” back, but you can only use it at Shoppers, which is ludicrously over-priced. So you get your 30-42% back in points (including the bonus redemption at the high end), but then you have to spend those points on stuff that’s 30-42% over-priced.

I had in my head that Shoppers was roughly 30% over-priced, but I hadn’t checked it explicitly in a while. So tonight when I went shopping, I did a price comparison on a basket of 11 items at Shoppers and Real Canadian Super Store (RCSS — basically Loblaws), chosen quasirandomly (I wandered around the store and wrote down the price of stuff I might find on my shopping list, and stuck to name brands so I could compare across stores). The average was actually that Shoppers was 42% more expensive than RCSS (data at the end).

So, even on the most fortuitous bonus points days at Shoppers (20X) you spend $1.42 to get $1 worth of stuff, and also pick up $.60 in Optimum points, which you later spend to get $.42 worth of stuff. In the end, you spent $1.42 to get $1.42 worth of stuff — but paid tax on $2.02. That tax hit would mean that you’d be behind by $0.078, or just over 5%. You go to all the trouble of shopping on the 20X the points days, then saving up your points for the top tier and a bonus redemption day, and then run around the store with a notepad and calculator so you’re sure to spend as close to $200 as humanly possible… and you end up 5% behind the ball. Could have just gone to RCSS and bought on any old day (and received a bonus 1% in PC points).

To be fair, Shoppers does put stuff on sale on top of the Optimum offers… but RCSS puts stuff on sale, too, so again it’s one less variable to worry about. You could do better if the Shoppers sale price is down to the RCSS sale price and you get some kind of bonus points event (either on the collection or redemption side), but will that be true of everything you put in your cart? Sometimes, yes: when I do stoop to shopping at Shoppers, it’s to scoop up the loss leaders that are actually cheaper than other stores, and that is usually all I’ll buy. Coke and facial tissues are perennial loss leader favourites at Shoppers — though as I said above, I can’t actually do that at the store near my house since they sell out (or, as far as I can tell, never carried any inventory in the first place).

The high percentage return of Optimum really masks what a mediocre deal it is, since everything is marked up to account for that Optimum payback (which means you’re doubly screwed if you ever shop there when you’re not getting bonus points, or if you forget your card).

* – the box of Huggies in Shoppers was a 72-pack, but RCSS didn’t sell that size. The closest was a 90-pack for $23.43, so I took the price per diaper, multiplied that by 72, and put it in to get a fair comparison to Shoppers.

Did you know that there are separate categories for patents? Odd ones include a separate category for flexible bags (almost 50 patent applications per year!), knots and knot tying (8 patents issued last year), railway mail delivery (zero patents since 1971 – it’s a wide-open market! Must think about patentable railway mail delivery technologies now…), button making (one patent every other year on average – think of the archaic buttons we must be using now), and baths, sinks, and spittoons (259 patents last year – somehow I suspect more for the first third of the category than the last third). Reading that, you may be thinking that there must be tens of thousands of categories to have such specific categories. While there are a lot of categories, this isn’t because of a ridiculous level of comprehensiveness — there’s only about a hundred or so.

I thought the guy with the drugs and bioactive compounds portfolio must have it rough with over 5000 applications last year, but telecommunications blows that out of the water with over 12,000 combined between telecommunications and mechanisms for multiplexing signal transmission.

Ordnance I had assumed would fall under military secrets, but managed to rack up over 100 [public] patents per year for the last three decades running. Unless that’s a misspelling of city ordinances, in which case yeah, there’s a lot of new ways to ticket people being dreamed up… though I didn’t think you could patent that.

Finally, 2011 seems to have marked a sudden end to the innovative efforts of the world’s farriers: consistently coming up with 5-20 patents even decades after the horseless carriage made the trade all but obsolete, suddenly 0 patents were issued in the field in 2011. Yes, even as recently as 2002, when Firefly showed us that space cowboys would still be herding cattle on horseback in the future, farriers managed to come up with 18 patentable inventions for shoeing horses.