Farmed Animal Funders Custom Shallow Review: On Selecting Funding Strategies In General And On Focusing Funding On Open-Access Scientific Research For Plant-Based Alternatives

By kierangreig🔸 @ 2019-01-24T19:11 (+33)

Some context: I currently work as an Analyst at Farmed Animal Funders. Farmed Animal Funders is a relatively recently formed group of funders who each give $250,000 or more annually towards charitable initiatives that support ending factory farming and moving towards a more sustainable food system. One of the things that I do in my work is writing ‘custom shallow reviews’ for funders to help them make donation decisions. The remainder of this piece is one of those custom shallow reviews with some identifying information removed. I think that there is value in posting this type of analysis publicly so that it can receive feedback and be updated accordingly, and so that it can potentially be used by other people facing similar problematics/questions. This shallow review was motivated by a funder asking to what extent they should prioritize funding open-access research for plant-based alternatives in their research.

-----------------------------------------------------------------

Selecting general areas to prioritize for funding and thinking about how to select them seems very complicated, and I feel uncertain about the best way of doing this. I am trying to make progress on these issues in a way that is useful, and in a way that can be incrementally built upon over time.

As always, please let me know if you would like me to go into more detail on anything in specific or if you have any questions.

Some Initial Thinking About Prioritizing Different Areas

I think it would be useful to briefly highlight two key approaches to funding allocation described by the Open Philanthropy Project:

1. The “default approach”:

1a) There is a common “set of criteria and processes” for all your funding,

1b) We can estimate how different funding options perform on that common
“set of criteria and processes,”

1c) Then prioritize the funding options that perform best.

2. The “diversifying approach”:

2a) There are some different “sets of criteria and processes” that somehow
can’t be compared on a common metric,

2b) You assign weight to those different “sets of criteria and processes” to
varying but significant degrees,

2c) You then reason to allocate funding across those “different sets of criteria
and processes,”

2d) Each of those “sets of criteria and processes” then allocates funding to
what’s best according to them.

A way of thinking about the “diversifying approach” that could be helpful is that if you were using this funding approach you would allocate funding to several different buckets, and then each bucket would use its own “set of criteria and processes” to decide how that funding is allocated. I think I aim to use ‘sets of criteria and processes’ in the same way that the Open Philanthropy Project describes worldviews. Alternatively, a “set of criteria and processes” might also be labelled as some “set of crucial considerations.” It could also be helpful to think of different “sets of criteria and processes” as different mindsets or different ways of thinking about how to prioritize between different funding options.

Some general “criteria and processes” (with examples further specified in endnotes)[1] that could differ between the different buckets within Effective Animal Advocacy (EAA) that you may want to allocate funding across are:

• Different buckets could have different processes for drawing inferences from evidence [2]
• Different buckets could have slightly different values that lead to large differences in how they think funding should be allocated [3]
• Different buckets could have different initial beliefs about the effectiveness of some class of interventions4 [4]

My reason for highlighting the difference between the “default approach” and the diversifying approach” is that I think a funding strategy of heavily funding (e.g., ≥80% of annual funding) plant-based alternatives can probably only come about if one is largely using the “default approach” and not the “diversifying approach.” If to a significant extent you think that the the optimal funding strategy is the “diversifying approach”, then I think that this would probably lead to a significantly greater split of your funding across different areas within and besides plant-based alternatives.

Further, I think that relative to the funding allocations that you [the funder for whom this report was initially written] seem to be considering (i.e., largely prioritizing plant-based alternatives), although I do feel very uncertain about this, would to a greater extent follow the “diversifying approach”. I think this approach would result in splitting funding across a few different prioritizing buckets to a greater extent, after which each of those buckets would allocate funding according to what seems best to them. At this stage, I seem to favor the diversifying approach to a greater extent than you. This could be because I may feel more significantly uncertain about how to prioritize funding options than you and/or I may struggle more with how to compare priorities suggested by different “sets of criteria and processes” to a greater extent. To the degree that you’re correctly confident in some specific “set of criteria and processes” for prioritizing funding options, the “default approach” does seem correct and I think that a plausible output of that approach could be to heavily prioritize plant-based alternatives with your funding.

I would be happy to elaborate on some of the reasons that might help explain why I currently tend to lean towards the “diversifying approach” to a greater extent than you, or why I think that following that approach to a non-trivial degree would lead to a significantly greater split across different areas than the one that you seem to be currently considering. However, I am unsure about how useful doing those things would be at the moment. Please feel free to let me know if you would like me to do either or both of those things. In addition, if you would like, I could further think about how you might like to apply the “diversifying approach” and suggest some possible “buckets” you may find appealing to divide funding between.

Some Initial Analysis Regarding Prioritizing Funding Open-Access Research For Plant-Based Alternatives

Plant-based options already seem to be:

1. (fairly) cost-competitive, though most still have significant room for progress
2. (fairly) taste-competitive, though most still have significant room for progress
3. stocked in some large supermarket chains
4. included in and selling out in(!) some large fast-food chains
5. comparing fairly well nutritionally to animal based foods,
6. are better for the environment than animal based foods.

My basic understanding from reviewing the overviews of 10 analyses of the global plant-based food market is that the plant-based market is now a several billion dollars in size and and seems to be growing at a rate of several hundred million annually. For instance, The Plant-Based Food Association’s estimate is that the US market is ~$3,3 billion, they estimated ~20% growth between mid-2017 and mid-2018, and the estimates from the market reports for the likely growth rate seem to put it between 5 and 10% CAGR [5] in next 5-10 years. Again according to PBFA’s analysis, plant-based milks now account for ~15% of total US milk sales and grew by ~9% in the past 12 months. So, to give a sense of possible scenarios in the future, I very roughly modelled an optimistic scenario of the plant-based foods excluding dairy, assuming that a growth rate of 20% per year until 2030 is maintained. This would bring the plant-based meat market to 10% of the global meat and seafood market if the rest of that market were to grow at 2% per year over that time period.[6]

On the other hand, I think that there’s much reason to be skeptical of some reported timelines for cost-competitive cell-based meat. Some reports—such as the Open Philanthropy Project’s Animal Product Alternatives report and van der Weele & Tramper (2014)—suggested that it is unlikely that cultured meat will become cost-competitive with conventional meat. [7], [8] I don’t think that New Harvest has responded well when asked about these concerns.[9] My impression is that groups initially give timelines that are too optimistic and then revise them to be more realistic timelines as it becomes clearer that those initial timelines were optimistic. [10] The market reports analyzing the cell-based market all seem to suggest that this market will likely be much smaller than the plant-based market in the near term (e.g., estimates suggesting that the cell-based market will total a few tens of millions but that the plant-based market will be many times greater than that, at a total closer to $10 billion). For those reasons, at the moment I would at least lean towards favoring the plant-based alternatives over the cell-based alternatives.

I think that it is reasonable to believe that one of either plant-based or cell-based alternatives would play a significant part in any longer-term massive decrease in the large-scale suffering from the industrial agriculture of animals. I think it is also fair to say that, for now, there is decent reason to prioritize supporting plant-based alternatives over cell-based alternatives. At this early stage in my research, the availability of taste and cost-competitive plant-based alternatives seems to be at least one of, if not the most likely credible longer-term avenue to massively decreasing farmed animal suffering.

I think in some ways, after settling on plant-based alternatives for animal products as a priority area, the case can be pretty quickly made for prioritizing supporting open-access research to plant-based foods. For instance, according to GFI:

The vast majority of commercially available plant-based protein ingredients comes from only 2 percent of the 150 plant species on which today’s global food supply depends. A significant pool of potential plant protein sources is thus available for exploration, and this does not even take into account the almost 250,000 additional plant species not used in agriculture today. Innovation opportunities in this area include expanding and diversifying our use of plant protein sources, determining which sources are best suited to particular plant-based meat products, and ensuring that the proteins from these novel sources are optimized specifically for plant-based meat rather than plant-based foods in general.

That seems to me to pretty strongly suggest significant knowledge gaps that could be addressed by open-access research. My quite basic understanding, based on very preliminary analysis, is also that some of the most significant gains in market share that the plant-based meat market has made, seem to have predominantly come from utilizing new processes or bringing different products—not just incrementally different, but quite different products—to market. For instance, one explanation that I have come across a couple of times but not really investigated due to time constraints is:

Originating from research in the 1990s, high-moisture cooking extrusion processes have enabled, in one step, the texturing of plant-based proteins into fibrous structures that mimic muscle-meat-like structures, giving them a bite and mouth-feel closer to meat. Perhaps the leading commercial example of this is the company Beyond Meat, whose products are famously said to have fooled the taste buds of Microsoft founder Bill Gates and others, leading to their investment in the company.[11]

That is, the prevailing interpretation seems to be that some open-access research [12] from the 1990’s played an important role in significant advancements in plant-based meats (this open-access developed process then being used in the Beyond Meat and Impossible Burger) around two decades later. If that is true, I think that could be significant evidence in favor of a funding strategy that were to prioritize open-access research for plant-based alternatives.

But to what extent are others already addressing this research gap? My rough impression is that thus far, there has been little interest in open-access research to plant-based alternatives. For instance, it was reported that GFI estimates that only about 0.3% of universities worldwide are doing research in the fields of plant-based and lab-grown meat. There seem to only be few programs, labs, or funding bodies that allocate funds to this issue. Based on a very preliminary search, the ones that do include:

• The European Commision’s Protein2Food
• The North Carolina Food Innovation Lab
• Plant meat matters
• Earlier this year, it was reported that the Canadian government announced that it will invest “up to to nearly $153 million, matched dollar for dollar by the private sector, in the Prairie-based Protein Industries Canada Supercluster.”
• Earlier this year, it was reported that Israel announced that they would invest $28 million into a food tech incubator over 8 years.

Through a very quick search (i.e., 10 minutes) for scientific articles completing research on plant-based meats I found ~7 (see endnote)[13] and then struggled to find more. So it seems like very little open-access research for plant-based alternatives occurs. In fact, if giving a few million to this area annually, one might actually double (or more) the amount of open-access research on plant-based alternatives within the next few years. This actually makes me somewhat concerned that open-access research for plant-based alternatives might not have adequate room for more funding in the near future. The risk is that it might not be able to absorb more than several million in funding without there being significant diminishing marginal returns to allocating further funding to it. At that point some other area within plant-based alternatives or EAA could emerge as the new option(s) that you should prioritize.

Now, I do wonder if a lot of research is happening within plant-based food companies. Plant-based food companies have received investment from several large funders, ~tens of Venture Capitalist firms, and a number of large corporate partners. Over the past few years these companies seem to have raised several-hundred million dollars in funding. According to Pitchbook Data, US plant-based startups raised $296M in just the last year. [14] For what it is worth, this degree of strong investor interest from others, I think, gives a lot of reason to be skeptical about the counterfactual value added by investing in these companies directly. My current opinion is that investing in plant-based food companies probably crowds out some other investor. This seems suboptimal because most farmed animal advocacy funders otherwise would likely use that funding relatively more productively and other investors seems less likely to do that.

I think it is fine to say—by a quick count there’s something like 94 plant-based food companies—that most of them are small-medium enterprises that most likely aren’t really allocating resources to the type of foundational research that I think you’re considering funding. My very rough impression would be that basically all these small and medium enterprise plant-based food companies are not investing much at all in research and development, and if they are, I would expect that it would mainly focus on incremental innovations around existing products and processes. [15] That type of research is likely quite different from open access research on new products and processes.

I am a concerned that some large companies would be completing their own research and development efforts that would to an extent be duplicated by the open-access funding. For instance, Quorn is reportedly spending millions of pounds on a new research facility and has a research team of more than 35 people. Beyond Meat’s research team is reportedly working on new products to follow up on the release their recently released sausage. According to GFI, JUST is now sharing a platform that is a “deep database of sustainable, functional tools from the plant and animal kingdoms” in order to launch a Food Technology Accelerator. I imagine that Impossible Foods would also have significant research plans. Other large food companies probably also have research and development, or product development that could be worth considering. For instance this FAIRR report profiles 16 of the largest global food companies, each worth billions of dollars, and several companies are referenced as having dedicated internal resources (R&D, product development, procurement) that focus on developing and/or acquiring plant-based products and ingredients. For instance, Tyson has reportedly been formulating vegan protein bowls that are set to hit stores in 2019. Some of these other large food companies could be acquiring twin-screw extruders to advance their own analogue efforts. Nestle has now released an “incredible” burger.

Unfortunately it seems these for-profit companies are incentivized to keep a tight lid on their research and as such, their research is almost always siloed within a company. As a result of that, I think it is reasonable to say that open-access research, even if it risks duplicating existing research in large private companies, still has high expected value because others would be able to more quickly build upon it.

I think that a deeper analysis of whether or not to prioritize this area would:

• Continue to examine this specific area of funding by looking at the extent to which open-access research from the 1990’s seemed to accelerate the plant-based market gaining of market share around 20 to 25 years later.
• Attempt to identify and do some initial evaluation of similar initiatives in comparable areas.
• E.g., perhaps there was or is an initiative regarding open-access research in fields like biofuels or renewable energy that would shed some light on the likely impacts of open-access research within plant-based alternatives.
• Think more about—and literally asking around about—what could incentivize some for-profit companies to make some of their findings open-access.
• Look in more detail into where would be best to allocate funding to open-access alternatives.
• For instance, I think that deciding what research to fund is very important. I would emphasise the need to be aware that at times there could be a large disconnect between what academics want to study and what would be useful for actually furthering the plant-based food market. For this reason it seems like a good idea to have GFI or other people with enough domain expertise involved in funding allocation. They would be able to tell when this seems to be particularly likely, in which case they could decide not to allocate funding to those proposals. However, I think that in addition to domain experts reviewing applications, what could be particularly useful is helping to set the priorities of any research fund. For instance, ensuring that research would apply to the most numerous and neglected farmed animals.
• Some other things to consider could be the complicated literature on the time-lag between the moment that open-access research and development is completed and the moment when it would pay-off. [16]
• What could be interesting to think about are the strategic ramifications to EAA if there were a significant influx in funding to open-access plant-based alternatives research now, and we would expect the pay-off of that to be in, say, 10-30 years.
• It could also be worth considering the complicated literature on the relationship between university research and private innovation, that I haven’t engaged with here. What I would note though is that the research on this relationship in the food sector seems to be scant.[17]

Initial Broad Overall Views of Priorities within Plant-based Alternatives

The reason that I haven’t continued to analyse open-access research to plant-based alternatives in more detail is I think that doing so is unlikely to significantly change my general overall view of how to best allocate funding within plant-based alternatives. My current overall view is that there are a number of competitive priority areas within plant-based alternatives, with the initial list as something like:

This link shows that same table with clickable links inside it.

In addition to funding open-access research of plant-based alternatives, other areas I might pick as as the ones that I find particularly promising:

• Providing Strategic Support for Companies or the Industry
• E.g., GFI reports they aided in launching Good Dot and DAO Foods International. In addition to GFI, a number of other non-profits are also looking to expand into this space and offer strategic support for plant-based alternative companies. This type of work could include helping raise funds for start-ups, connecting co-founders, and consulting on marketing practices.
• Trade associations could also meaningfully support the entire industry.
• Preventing Unfavorable Regulation of Alternatives
• See lawsuit from several groups, including GFI and the Animal Legal Defense Fund. countering the first US law (in Missouri) legislating that plant-based products cannot be called meat. The Plant Based Food Association (PBFA) for instance is also looking to scale up in this area.
• Supporting plant-based alternatives in important lower and middle income countries.
• The Asia Pacific region has seemingly created few plant-based companies but this is where the market is predicted to have the greatest relative growth in the near term. I wonder to what extent GFI or others are planning to address this, I may reach out to them about that.
• It seems worth noting that the animal product alternatives landscape in these in lower and middle income countries is much more neglected and less developed than it is in high income countries.
• Further consumer acceptance research
• E.g., One report suggests that “plant-based” has no equivalent term in asian languages. I wonder to what extent GFI or others are planning to address this, I may reach out to them about that.
• Campaigning to Stock Plant-Based Options in Accessible Fashion at Major Selling Points

I think that I would overall describe the current situation of choosing which general areas to prioritize in plant-based alternatives as follows:

1. There are a number (i.e., > 10) possible general areas
2. There is much uncertainty about those general areas
3. Based on some initial analysis some of the general areas, in expectation, appear much better than some other general areas
4. For me, within those general areas, (i.e., those that seem better than some other general areas and are listed in the above bullets), I currently don’t think that I can say with high confidence which is best.

That is, when I spend time reflecting on those the bulletted areas listed above, to me there don’t really seem to be strong reasons to now think that, in general, supporting one of those options with several hundred thousand dollars or more annually is better than supporting another with an equal amount. I think that I could be convinced that there are strong reasons to think that, but that would require much further analysis. With the information that I have, I only really lean towards some area as being more effective than another area in general. That is, if I do a head to head comparison between any two options from that bullet listed above I am not sure that I can currently assign more than, say, a 65% probability that funding one of those options with several hundred thousand dollars annually for the next two years is more effective than funding another option with that amount of money over the same time period.

If pressed, at the moment I am not sure that I would say that funding open-access research to plant-based alternatives is the top priority area within plant-based alternatives, but I don’t think other areas are clearly better than funding open-access research. Something that makes selecting which general areas to prioritize complicated for me is that the variance within any of the bulleted possible priority areas seems much greater than the variance between them. So, for instance, some of the less-than-average funding options within the area that seems most promising can be less promising than some of the greater-than-average funding options in some other area. Supporting promising campaigns to stock plant-based options in accessible fashion at major selling points, or supporting plant-based alternatives in important LMIC are areas that I think are now the most likely to outperform open-access research to plant-based alternatives. However, I don’t have high confidence in that. Moreover, there would be different ways of supporting those general areas which are less promising than the better ways of supporting a number of other promising general areas within plant-based alternatives. Please let me know if you would like me to expand on the reasons for why I think this.

I think that would imply that if you are wanting to fund the most promising options within plant-based alternatives, then rather than heavily allocating funding to one specific area, it seems that a better approach would be to fund the outstanding opportunities from a few different priority areas. I think that if I were allocating a few million dollars to several million dollars per year to plant-based alternatives, I think that at this point and into the near future, it would be unlikely that I would allocate more than, say, 70% to open-access scientific research on developing plant-based alternatives. All that said though, I do want to stress that I have low confidence in my analysis at the moment. I also significantly updated about this area in the short time that I spent looking into to it, and if I were to look into it further I may continue to significantly update about it.

Main Tentative Conclusions And Suggestions From This Shallow Review

• I do feel uncertain about this but I think I would tentatively suggest that you should consider implementing, to a greater extent, a “diversifying approach” to your funding allocation.
• I think that if you were to implement the “diversifying approach” to any significant extent, it is likely you would allocate significant funding across at least two different general priority areas (e.g., plant-based alternatives and at least one other area) and plant-based alternatives probably wouldn’t receive >80% of your annual funding allocation.
• If allocating significant funding to plant-based alternatives (e.g., 80% of your annual funding amount) my current general impression is that it would probably be best to split that across multiple outstanding opportunities within the plant-based alternatives sphere rather than just focusing that on open-access research to plant-based alternatives.
• My current weak and general impression is that if using several to a few million dollars to try and fund the best options within the plant-based alternatives sphere, I would say that the optimal annual allocation is currently unlikely to result in more than, say, 70% of that funding to open-access research on plant-based alternatives.

Thank you Rethink Priorities and Jay Quigley for providing feedback on this piece.

Endnotes:

[1] Please note that this is some of my initial thinking about this idea. My examples given in the endnotes below are more like some plausible candidate buckets for splitting funding across rather than those I might suggest if I was to think about this more, and the percentages assigned to them are for now basically just for illustrative purposes.

[2] When I think about this, a possible example in our context could be that in addition to the primary bucket that could be allocating most of your funding to plant-based alternatives there could be another bucket with, for instance, ~10% of your annual funding. This other bucket requires a relatively large amount of positive evidence about the effectiveness of some funding option before allocating funding to it, and if it can’t find such a funding option then it allocates its funding to further research in order to determine those funding opportunities in future. All other buckets may disagree with that approach to handling uncertainty or evidence and may be more comfortable allocating funds in light of more significant uncertainty about the effectiveness of a funding opportunity.

[3] When I think about this, a possible example in our context could be that in addition to the primary bucket that could be allocating most of your to plant-based alternatives there could be another bucket with, for instance, ~10% of your annual funding. This other bucket could have values that suggest that most wild animals have lives that are significantly filled with suffering. Those values may then imply that donating to groups that focus on reducing wild-animal suffering is the priority and that bucket allocates its funding to the best funding opportunities in that area. All other buckets may disagree with prioritizing that area.

[4] When I think about this, a possible example in our context could be that in addition to the primary bucket that could be allocating most of your funding to plant-based alternatives there could be another bucket with, for instance, ~10% of your annual funding. This other bucket could one bucket that has an overall worldview that suggests that the set of interventions that are say more traditionally associated with social movements (e.g., the set of corporate campaigns, legal initiatives, and veg*n advocacy) as opposed to plant-based alternatives will be the more effective option to fund. This bucket could then attempt to find the very best funding opportunities in that other set of interventions and attempt to fund them. All other buckets could disagree that this set of interventions is going to be the most effective.

[5] For instance, this estimate of 9.4% CAGR for Asia Pacific region and a 7.7% CAGR for the entire plant-based market. The Asia Pacific market is reportedly less than $1 billion and so growing at perhaps an estimate ~$60-70 million annually.

[6] Please let me know if you would like to see the spreadsheet that I did this calculation in. I would then tidy it up and share it with you.

[7] “We are highly uncertain about the eventual cost per kg of cultured meat, and have not closely examined the above cost estimates. However, none of these estimates suggest a cost competitive with that of conventional meat.”—The Open Philanthropy Project (2015). Animal Product Alternatives. The Open Philanthropy Project.

[8] “From an economic point of view, however, competition with ‘normal’ meat is a big challenge; production cost emerges as the real problem. For cultured meat to become competitive, the price of conventional meat must increase greatly.”—van der Weele, C., & Tramper, J. (2014). Cultured meat: every village its own factory? Trends in Biotechnology, 32(6),294-296.

[9] “While it is possible that cultured meat may never reach cost-competitiveness, New Harvest says that this doesn’t make the research they are supporting worthless.29 New Harvest likens the situation to biofuels—biofuels aren’t cost-competitive today for many reasons but one largely being that they compete with an artificially-priced commodity: fossil fuels. While we aren’t currently using biofuels everyday, in the event of a catastrophe or growing fuel prices, that technology could be mobilized. They think this situation is similar to that of cultured meat. New Harvest also claims that they haven’t done enough research in the area to suggest that cost-competitiveness is an impossible task. They do note that cost-competitiveness is difficult, but they don’t yet know for sure that it’s impossible.”—ACE’s 2017 Review of New Harvest

[10] For instance, in January 2016 New Harvest reported that acellular products would be available within a few years. Now, a few years later, they say they have no formal predictions about when products will become available.

[11] This quote is from here. Another example of this could be found in this article: “‘In the last 10-15 years, the quality of the product has improved tremendously.’ Much of that improvement can be attributed to his adoption of twin-screw extrusion technology and its ability to align the fibers in plant proteins.”

[12] This article seems be the one often cited as the research which lead to the advancements.

[13] Here are the links:

https://www.sciencedirect.com/science/article/pii/S0023643817307041

https://onlinelibrary.wiley.com/doi/abs/10.1111/jhn.12546

https://www.sciencedirect.com/science/article/pii/S0959652618320080

https://www.sciencedirect.com/science/article/pii/S2213329118300911

https://onlinelibrary.wiley.com/doi/abs/10.1111/ijfs.13847

https://onlinelibrary.wiley.com/doi/abs/10.1002/jsfa.9438

[14] H/t Lewis Bollard who cited this statistic in a 2018 newsletter.

[15] For instance, ‘[t]he food industry is classified as low-tech on the basis of low R&D investment and low levels of human capital. For this reason, it is believed that these firms have low capabilities of innovation and that they mainly focus upon incremental innovations around existing products and processes (Galizzi and Venturini, 1996; Grunert et al, 1997).”

I also found that this 2018 technical report for the European Commission was helpful in forming a rough impression.

[16] See, for instance, Ralston (2010)

[17] For instance, “[t}he literature on the relationship between university and innovation is extensive and rapidly growing, but research on the food sector is scant.”

[18] See previous shallow review on Sutardja Center for some further reasoning about why to prioritize these universities and specifically the program at Berkeley.