Growing the US tofu market—a roadmap

George Stiffman20 Oct 2022 16:46 UTC
284 points
37 comments18 min readEA link

This post is a roadmap to growing the US tofu market. There is a ton of delicious impact to be had, and I hope more of you in the alt protein space will consider joining me.

(Thank you to Kris Chari and Nick Corvino for helpful feedback.)

A yuba-based Napoleon pastry from my forthcoming book, Broken Cuisine[1]. Photo credit: Ryan Tanaka

Why should I care about tofu?

When I first visited China, I assumed tofu was just a pasty, white block.

I never would’ve imagined that tofus could melt… or that some varieties tasted like aged cheese… or that others had a bread-like crumb!

Chinese tofus are some of the most culinarily exciting plant-based proteins in existence. Yet, hardly anyone outside the Chinese community knows they exist!

Yunnan juicy tofu, grilled over charcoal with seasoned chili powder.

The average American eats just 1 block of conventional tofu per year[2]. Yet, even this tiny consumption, added up, is enough to displace millions of animal lives from factory farms each year[3].

While I personally love firm, soft, and silken tofu, these ingredients are poor fits for western cooking styles and taste preferences. In contrast, several Chinese tofus are much, much, much better fits[4]. If we can find the “avocado toast” of these tofus, expand ingredient supply, and successfully promote, the market for these varieties could thus be many times larger, someday saving 10s or 100s of millions of animal lives each year.

Beyond directly reducing demand for meat, promoting Chinese tofus could have several other benefits:

  1. Chinese tofus could make plant-based cooking more appealing to chefs in ways that meat analogs (plant-based meats, cultivated meat, etc.) cannot[5]. If true, this could turn some of our biggest opponents to plant-based progress into allies and evangelists, shifting the culture [6]around dining. Shifting culture could have more impact than merely substituting meat for mock meat or tofu[7].

  2. Better tasting vegan food seems necessary for moral circle expansion[8]. Chinese tofus are one of the best opportunities to expand vegan cooking, as their unique culinary properties can do things that meat and vegetables cannot[9]. (This seems especially pressing if you have short AI timelines and worry about value lock-in, but there don’t seem to be many interventions currently being pursued.)

  3. Making tofu a high prestige food in the States could possibly raise its status in China. After living in China for several years and studying the Chinese farmed animal welfare space throughout 2020, I am pretty pessimistic about near-term prospects for animals. Promoting tofu in the west, to raise its status in China, could, however, be a straightforward and significant way to contribute.

  4. Because traditional plant protein advocacy is so neglected within the alt protein space, there could be a lot of information value in trying to promote Chinese tofus[10].

Beyond being massive in scale and very neglected, I am hopeful that this opportunity is tractable, as you’ll see below...

Wait! What is Chinese tofu?

China is the birthplace of tofu and has over 20 distinct varieties[11]. While some of these varieties are common across Asia, the majority are specific to Chinese cooking. There are a few, in particular, that seem to be great fits for western cooking styles:

  1. Shanghai tofu (上海素鸡 shanghaisuji) - ultra high in protein, rich/​eggy/​custardy flavor, bready structure that can be cooked into cakes, soup dumplings, “crostini”, protein crumbles, etc.

  2. Spongy tofu (千页豆腐 qianyedoufu) - a fishcake-like tofu, pre-seasoned (NOT bland), delicious blended, braised, or grilled, can be dense and chewy or light and silky.

  3. Fermented tofu (腐乳 furu) - a funky, cheesy, umami seasoning that’s blended into sauces or rubbed onto bread or rice.

  4. Pressed tofu (豆腐干 doufugan) - a type of extra-firm tofu, wintery-spiced flavor, great on grilled cheese, grain bowls, with apple/​sugar/​cinnamon spices, etc.

There are a number of rarer tofus that I hope can one day be available in the west, but that are already hard to find in China:

  1. Juicy tofu (包浆豆腐baojiangdoufu) - a meltable tofu popular in Yunnan province.

  2. Charcoal ash tofu (荞灰豆腐 qiaohuidoufu) - a tofu from Guizhou province that’s coated in ash and slow cooked for a few days. It looks somewhat like pressed tofu, but has a very tender texture and nice smoky flavor.

  3. Bubbly tofu (泡豆腐 paodoufu) - a thin tofu sheet from Guizhou province that’s dried then cooked in hot rocks to puff up. It needs to be reconstituted (soaked) in liquid before using and has a wonderful texture.

Another great-fit tofu is yuba (油豆皮 youdoupi) - the film that forms atop heated soymilk. It’s high in protein, keto, rich in flavor, easy to season, and has many interesting uses—from laminated pastries to easy protein add-ons for salads or pasta. I don’t include it in the above list because it’s eaten across Asia so is not strictly Chinese.

A smorgasbord of Chinese fermented and pressed tofus. Photo credit: Ryan Tanaka

How do we popularize these foods?

Our first challenge is to create a market for Chinese tofus for non-Chinese audiences. To do so, we need to solve three bottlenecks:

  1. There’s very limited supply.

  2. Because of limited supply, most Americans aren’t aware that these tofus exist.

  3. Because there is limited supply and awareness, there’s little exploration into how to use these ingredients in other cuisines beyond Chinese cooking—meaning it’s unclear what needs they’re solving.

Our second challenge is to scale this market as big as possible. To do so, it would be helpful to understand a few questions:

  1. Historically, how have new foods become super popular?

  2. How can food cultures learn from each other without culturally appropriating, or being perceived as culturally appropriating?

  3. How can we get the public to re-embrace soy?

I want to emphasize that few of these things will happen without more firepower. There is so much to be done in this space—so much delicious and meaningful impact to be had! If you’re even slightly tofu-curious, let me know here. See the summary at the bottom of this post for a long list of ways to support.

Creating a Market

Supply.

Current supply.

There’s very little Chinese tofu in the States.

In fact, there’s not a single company here specifically focused on promoting Chinese tofus. This is probably because the market is tiny and importing frozen goods at low volumes is expensive. As just one example, I estimate current annual profits from importers of Shanghai tofu, or suji, to be on the order of $10,000/​year[12]. That’s not even enough profit to support one full time employee. Other Chinese tofus fare similarly[13].

For low volumes, cold shipping is prohibitively expensive. There are two main options: air freight and ocean freight. Ocean freight is relatively affordable, but requires more volume. I estimate that the smallest ocean container could fit roughly the entire US annual supply of Shanghai tofu[14]. Because several trading companies already compete for that market, an extra container of product would likely sit in storage. On the flip side, air freight has 10x lower minimums in terms of container size[15], but the unit cost per tofu product is way too high, likely over $5/​log or block-worth[16]. Of course, you can try to book a portion of an ocean container, but that’s both a hassle and more costly.

The main workaround is to bundle small quantities of various products into one shipping container, then ship via ocean freight. By using a whole container, trading and import companies can secure much better rates. This is where most of the US supply comes from.

A limited supply of a couple other tofus—pressed and yuba are—are domestically produced. I’m familiar with at least four brands producing pressed tofu, and at least one making yuba. These markets are slightly larger—pressed tofu is used across Chinese cuisines, so likely has a 10-100x bigger market than Shanghai tofu; some varieties of yuba are used in other Asian cuisines, making that market bigger too. (That said, a $1M market is still very small, especially when multiple producers and importers are competing for it.)

These ingredients are available, in limited supply, through Asian food distributors and Chinese supermarkets. Occasionally, a Japanese, Korean, Vietnamese, or pan-Asian market might carry a couple Chinese tofus, but this is uncommon, and the selection is very limited. There is also at least one online grocery store that carries these products. (www.sayweee.com)

Building supply.

There seem to be three ways to grow the US supply of Chinese tofus.

First, we could encourage existing trading companies to import more tofu and help them partner with more mainstream distributors, maybe by working from within the company or acting as a sales broker. I’m not excited by this approach, for a few reasons:

  • Distributors won’t pick up new ingredients unless there’s clear demand. Yet, it’s hard to build demand for an unfamiliar product without spending a lot on marketing, and these trading companies don’t have sufficient budgets.

  • These trading companies aren’t investing much employee focus into their tofu lines.

  • They generally haven’t responded to my outreach, though I could be more proactive with follow-ups.

Second, we could help local producers scale up. This might involve supporting a company that already produces the product, or encouraging a conventional tofu company to add new product lines. In terms of current producers, only pressed tofu and one variety of yuba are manufactured in the States. While most of these products (in my mind) are lower quality than the imported stuff, there are a couple good ones. Convincing a brand like Huaxing to invest more into marketing and mainstream sales channel could be valuable; granted, as with supporting trading companies, the biggest challenges here are validating demand and securing funds for marketing. I’m not sure how possible that is.

To convince manufacturers to produce more of the other Chinese tofu varieties, we would again need to help them validate demand. Yet, this runs into the same problems around marketing spend and employee attention, while also requiring capital investment on new machinery. I’ve chatted multiple times with the head of Hodo Foods, the third leading US tofu producer, about adding other tofu products, and they’ve been disinterested each time.

Third, we could help start a new distribution company specifically focused on sourcing and promoting tofu. This would mean finding reliable factories in China, handling import logistics, and selling to US distributors. This is the approach I’m most excited about and is what I’m currently working on. A key advantage is that it could raise venture funding, which could pay for marketing spend and a full team. It could also give us pricing power to keep tofu costs low.

Starting an import company would require a few things:

  1. Finding product-market fit. If we can show that a specific tofu fulfills a specific market need, and that market is large, then we can demonstrate to investors that there is a large business opportunity. To get there, I’m trying to persuade LA restaurants to pilot tofu menu items within specific niches (i.e. tofu sheets as Italian “pasta,” spongy tofu as the center of a braise, pressed tofu slices in grilled cheese, etc.)[17] If restaurants see benefits, then I’ll pitch to similar restaurants, collect an interest list, and show that to investors. If the pilots don’t work well, then I’ll try to figure out why and adjust[18]. While I just started restaurant outreach and am unsure how many warm leads will convert, I’m hopeful.

  2. Raising funds. Building a pitch deck and business plan. Finding partners – co-founders, Chinese tofu factories, promoters, distributors, storage facilities. Incorporating as a c-corp. Building branding and basic social media/​web presence…

  3. Importing a container of one type of tofu. Pitching our products to distributors. Selling. Marketing. Invest in R&D to find more use cases.

  4. Scaling out to other varieties of tofu.

Ways you can support:

  • Co-founding a tofu import company. I’m looking for a co-founder with some of the following characteristics: Chinese or Chinese American, action oriented, knowledgeable about startups, social media, or the food industry.

  • Funding.

    • I could use grant funding to cover my cost of living before raising venture money. This would allow me to work on this full time, maybe accelerating progress by 3-4 months. (I’m applying to the EA Animal Welfare Fund but am unsure if my grant will go through.)

    • I’m also looking for angel investors to invest in our seed round. Not immediately, but once we validate the market, hash out our business model, and have a better sense of budget needs.

  • Advising. On the food side (importing, distribution, restaurant sales) or business side (fundraising, building startup infrastructure).

  • Offering connections. To potential co-founders, advisors, funders, or other food people (chefs, restaurateurs, influencers, food writers, culinary school directors, folks who work with restaurants, …)

  • Volunteering. Helping with informational graphics, photography, import company branding, …

Few use cases and low awareness.
(Discussed together, as their solutions overlap.)

An adapted bagna cauda dip made from shiitakes and fermented tofu, rather than anchovies. Photo credit: Ryan Tanaka

In western cooking.

To create a market for Chinese tofus in the States, we need to develop scalable ways to use it. By scalable, I mean uses that could be adopted quickly by a wide range of food service establishments. Scalable uses tend to be delicious, low-cost, and easy to assemble—things that could become the next avocado toast, freezer waffles, or acai bowls. For these ingredients to take off outside of the Chinese American community, the use cases will likely be within western cuisines.

There are several ways to create more use cases.

  1. Hiring an in-house chef . A couple years ago, it might cost $5-20k to hire a well-respected R&D chef to develop 20-30 recipes, targeted towards specific use cases. A tofu importer has the clearest case for investing in such R&D; they could use recipes to market to new clients. That said, recipes could also be commissioned by others interested in promoting alt proteins, or by organizations that encourage restaurants to veganize their menus, like Veganuary. They could also be commissioned for a consumer-focused cookbook—but this seems like less of a priority[19].

  2. CPGs. Instead of developing uses to ‘cook’, someone might develop recipes for a CPG (packaged food) brand. Doing CPG R&D wouldn’t be as helpful for promoting tofu “as an ingredient”[20], but it could generally help promote plant-based foods. Moreover, if targeted towards food niches dominated by chicken/​fish/​eggs, it could scalably reduce animal suffering.

  3. Out-of-house R&D. Instead of hiring a chef, we could inspire food bloggers, restaurant chefs, or CPG brands to develop their own recipes. If we mailed ten $100 tofu boxes to ten different food bloggers, I’d guess that at least one or two would try out, like them, and create a video or two on their channel. While it would be harder to direct their R&D towards the highest impact food niches, it’s likely that there would be overlap. More significantly, these videos could expose hundreds of thousands of subscribers to Chinese tofus, creating instant hype. With a little coordination with the bloggers, we could direct this hype in a few ways:

    • To persuade grocery stores to carry these ingredients, we could add a petition to their videos → collect names and zip codes → send petition to nearby restaurants or grocery stores → connect those restaurants/​grocery stores to distributors.

    • Even better, we could add a mailing list signup. Collecting email addresses would allow us to send future asks, share new ingredient availability, or promote anything tofu.

    • We could show CPG (packaged food) companies the interest in those videos and try to persuade them to start tofu CPGs.

    • We could create paid ads to market those videos more widely and increase their reach even further…

One risk of this approach is not being able to curate the viewer experience. If a food blogger introduces these tofus to hundreds of thousands of viewers in a suboptimal way, that could lock in disinterest[21]. That said, this risk can be minimized by working closely with bloggers—offering recipes and general support—or by selectively screening potential partners.

How much impact would this have?

Developing novel uses cases expands the total addressable market for these ingredients. Finding a use that’s 10-100x more scalable than the current uses (i.e., in Shanghai or Beijing cooking), could literally expand that total addressable market by 10-100x. (This wouldn’t expand the actual market 10-100x; scaling requires a lot more magic. But it would make it easier to raise investment for a venture, find partners, etc.)

Five-spiced pressed tofu in a coconut dulce de leche apple tart from my forthcoming book, Broken Cuisine[1]. Photo credit: Ryan Tanaka

In Chinese cooking.

Besides creating new uses for Chinese tofus in western cooking, there would be a lot of value in highlighting traditional Chinese uses. This could build demand for these tofu ingredients, while raising awareness for vegan Chinese food, making it easier for meat lovers to find delicious alternatives. Some paths to impact in this area include:

  • Partnering with Chinese food bloggers. Linking petitions or mailing list signups, as described above, could encourage distributors to carry Chinese tofus, improving supply. (Granted, it might be less useful for creating new demand, if Chinese food blogger audiences are already familiar with the ingredients.)

  • Writing a book. Great books seem to have been instrumental in establishing their own culinary fields and making them widely accessible[22]. I wonder if writing a bestselling, foundational cookbook on [veg] Chinese tofu cooking could establish it as a high status culinary art, something up there with cheese, soul food, or sushi. If that were possible, it could create a lot of demand for these ingredients and encourage others to throw a lot more resources at starting restaurants, researching these foods, or educating others about them. Assuming there were good fit people who wanted to pursue such a project, investing $100k with a 5% chance at a breakaway bestseller seems like it could do more good than investing similar money in the alt protein space[23]. The cost to produce a book would be much higher than one-off partnerships[24], but the upsides seem high.

General demand.

Another option to grow demand would be to open a tofu-themed restaurant tailored towards a specific niche (taco fillings, tofu in Chinese cooking, fast food, etc.) Several upsides to this approach:

  • Restaurants that develop cult followings have incredible marketing power—able to influence other restaurants, consumers, media, etc.

  • A ghost restaurant model, in particular, could quickly validate different menu concepts. Some of these could be scaled to brick and mortar locations or sold/​marketed to others. This would not be too expensive to set up[25].

  • A restaurant that went through significant volumes of tofu could create a lot of demand for importers, improving their early stage cash flow.

  • Restaurants are the best way to curate customer experience and give consumers a positive first impression of tofu. That said, this is tough and slow to scale.

Ways you can support:

  • Co-founding the import project.

  • Founding a CPG brand. (I can share a long list of product ideas and connect you with tofu suppliers.)

  • Funding me

    • $10-20k to commission an R&D chef to develop scalable tofu use cases.

    • $1000 to send tofu boxes to 10 food influencers.

  • Writing a bestselling book on Chinese tofu cooking.

  • Partnering on social media promotion.

  • Opening a ghost restaurant.

  • Connecting me with chefs, restaurateurs, influencers, food writers, culinary schools, …

Scaling

Understanding how ingredients become super popular.

A number of ingredients have recently taken off in the States. Researching how these foods became popular could help us be more effective in promoting Chinese tofus, or other plant-based ingredients.

Some foods that could make interesting case studies include:

  • Kimchi

  • Gochujang

  • Sriracha

  • Salmon

  • Sushi

  • Bubble tea

  • Oatmilk

Some relevant research questions might include:

  • How fast did these foods become popular? What did that growth look like over time? Was growth generally constant, or did it follow discrete stages?

  • Who were the main interest groups that promoted these foods?

  • What environmental conditions supported this growth?

  • What challenges did these foods overcome, and how did they overcome them?

  • What challenges have they still not solved?

  • Did these ingredients take off via similar pathways? If not, which pathways seem most relevant to Chinese tofus?

I’m sure there are many other relevant questions we could investigate.

How much value would this research provide?

I think it’s unlikely that a new market could be created without certain elements: an early adopter community, scalable use cases, stable ingredient supply, etc. That said, scaling such a market is much more uncertain. Great strategies could have 10 or 100x more upside than average strategies (i.e., a 10-100x greater eventual market size.) Some strategies can also operate much, much faster than average ones.

Worst case scenario, research findings don’t generalize. Mid-case scenario, research simply teaches us what we would have learned from iteration, mentorship, consulting, etc. – accelerating our progress by several months or years. Best case scenario, research improves our odds of success and helps us grow the tofu market 1x, 2x, or 10x more than we originally could have.

The impact would thus equal the new tofu trajectory minus the old tofu trajectory. Say research simply accelerated progress by 6 months, with a 10% odds of eventually doubling the tofu market: that would mean nearly 250,000 factory farmed lives averted[26]. If you think (as I do), that the tofu market could grow 10x, not just double, then that number increases proportionally. Or assume, instead, that we had 10% odds of doubling the tofu market and better research increased those odds to 15%. In expected value terms, that could mean nearly 250,000 factory farmed lives saved each year[27]. What if we had a 10% odds to double the tofu market, but research gave us a 10% odds to triple it? That would mean nearly 500,000 lives saved each year[28]. Since research might affect all of these factors, the impact could add up. (These calculations are obviously very simplistic. Percentage growth in the overall tofu market is a bad benchmark—much better to look at individual use cases and the addressable obtainable markets within[29]. Yet, it’s hard to do so without more clear use cases, thus the need for culinary research.)

This is all to say, accelerating change could have a lot of benefit, but it would be even more impactful to increase our odds of growing the market, or growing it to much larger size. A month of researcher time (or less) could have a large impact.

Ways you can support:

  • Conduct this research.

  • Mentor this research project.

  • Connect me with someone who might want to pursue this research.

Avoiding cultural appropriation.

A failure mode I am particularly concerned about is being perceived as culturally appropriating tofu, which could alienate important allies and make it harder to grow the tofu market in the future.

To be honest, I didn’t really understand this worry until recently. I lived in China for several years, have Chinese god parents, and worked in tofu production. Tofu artisanry is a low-class trade in China, and my shifu (teacher) was ecstatic to meet a foreigner who wanted to promote his craft abroad. Most other Chinese people I meet are supportive too – what better way to counter rising anti-Chinese sentiment than delicious food, that’s also healthy and great for the climate?

Yet, back in the States, I’ve encountered some genuine worries about cultural appropriation, which seem to boil down to one thing: historically, white people have commodified elements of cultures they weren’t born into, profited from it, and left the traditional cultures stuck worse off and less understood – often needing to fight through new racial stereotypes. While it’s unlikely that people in China would suffer from a white guy trying to grow the US tofu market, perhaps Asian Americans might, if done carelessly.

To that end, I think it’s important that this work doesn’t inadvertently harm the local Asian American community, such as by misrepresenting Asian American tofu culture, diminishing or crowding out existing tofu voices, implying cultural ownership, or anything like that.

Beyond avoiding the direct problems with cultural appropriation, it’s also important to avoid its perception. Alienating Chinese American, Asian American, or young liberal allies would make it much harder to grow the tofu market. As described above, merely delaying progress by 6 months could lead to 250,000 lives of horrible factory farmed suffering. If opposition to this work spread beyond Asian American communities, and tofu became politicized, that could permanently curtail its future demand; the toll would be millions of animal lives. If we can lower these odds by even 5%, the impact could be millions of animal lives saved.

Here are some concrete steps that could reduce this risk:

  • Finding other Chinese Americans to collaborate with. If you are Chinese American and excited about popularizing Chinese tofus, I would love to chat. If you know of folks who are in that boat, please pass them my contact info!

  • Performing a risk analysis to figure out the worst ways we could go wrong, including looking at case studies.

Ways you can support:

  • Co-found or partner on one of these projects (for Chinese or Chinese Americans.)

  • Advise around cultural appropriation-relevant risk, branding, or PR.

  • Help run a risk assessment.

All in all, I don’t think cultural appropriation is as big an issue in food as other spaces of society. No one boycotts Thai food for using American chilies, or Mexican food for using European wheat flour tortillas, even though those ingredients originally came from a different culture. Let’s find a way to do this right.

Destigmatizing soy.

While I don’t think these beliefs are as widespread as most people assume, some Americans are wary of soy, believing that:

  • Soy can disrupt hormones, cause infertility, give men boobs, etc. (This is maybe possible if you eat obscene amounts of soy, but not if you eat even a block or two a day[30].)

  • Soy is a common allergen. (False – just 0.1-0.6% of Americans are allergic. Compare that with 2-4% for dairy[31].)

  • Soy is bad for the climate. (False – most soy is grown for animal feed, which is horrible for the planet. If soy were eaten instead of meat, it might reduce land use by 9x[32], which can create space for reforestation, if you’re into that[33]. One reason why soy is so widespread is that its hyper efficient to grow.)

  • Soy is largely GMO. (True – but this isn’t a bad thing. It may even be good for the climate as GMO soy requires less land. Also, most of GMO soy is fed to livestock – if you’re afraid of GMOs, you should be scared of meat.)

If soyfoods were less stigmatized, I think the tofu market could be much bigger. One scalable source of growth might come from the CPG space. A ton of new food and beverage brands are launched each year, and they absorb millions in investment and invest millions into marketing and consumer education. Yet, more and more brands are moving away from soy. Making it easier for new CPG brands to use tofu could a) funnel more money (from CPG brands and their investors) into tofu marketing efforts, leading to increased overall tofu demand; b) create more plant-based food options for consumers, with a portion replacing meat demand, depending on the type of CPG.

Another source of demand could be from the bodybuilding community, if reducing misconceptions around soy and testosterone led more bodybuilders to explore soy protein (rather than, say, chicken breast).

Destigmatizing soy for alt protein brands might allow them to use cheaper protein, lowering their costs[34]?

I’m unsure how to begin quantifying that potential value. There are probably many sub-communities that could use more tofu instead of animals. The tofu market could probably grow bigger in general.

I’m also unsure how tractable it would be to reduce stigma. Some half-baked strategies might include:

  • Running a campaign to establish anti-soy attitudes as culturally insensitive or racist. The argument being, soyfoods have been consumed for thousands of years in Asia and there is no evidence that they are unsafe for humans at normal consumption levels. Thus, being anti-soy is ignoring Asia. A campaign could include a series of video ads, featuring Asian American celebrities or doctors and various big brand partnerships. (I don’t think people who are anti-soy are racist – but convincing a swath of Americans that being anti-soy is culturally insensitive could be one way to reduce stigma.)

  • A media campaign pointing out double standards between how consumers view edamame (fresh soybeans, seen as a health food), miso (a soybean-based seasoning), and tofu. I genuinely believe there is a double standard between how Americans view Japanese-branded foods (refined, high end, healthy) and Chinese-branded foods (cheap, mass produced, fast food). Such a campaign should be done in a way that doesn’t put down Japanese cooking, but instead celebrates both.

  • A partnership campaign with leading and respected health agencies clarifying misconceptions around soy.

  • A media campaign making fun of the ridiculous soy boob myth and leading into more educational content. Perhaps featuring people perceived as healthy and fit, like professional athletes.

An alternative to directly reducing soy stigma would be to remove the elements of soy foods that consumers most worry about—isoflavones, the phytoestrogen in soybeans. Removing this component entirely might alleviate fears of hormonal disruption and make it possible to eat even larger quantities of soy protein. The first step might be to conduct a literature review on factors that effect isoflavone concentration in soybeans, such as varietal[35], cooking time[36], processing[37], etc. Perhaps there’s a cheap, efficient processing method that can remove this content from tofu. Or perhaps there’s a type of tofu with naturally low isoflavone content.

(Granted, this seems like a risky strategy. It feels likely that marketing foods as non-GMO reinforces consumer concerns, leading to greater anti-GMO sentiment. Something similar could happen here, increasing opposition to mainstream soy foods.)


Ways you can support:

  • Organize a destigmatize soy SWAT team.

  • Conduct a literature review on how to lower the isoflavone content of soyfoods or tofu.

  • For an MD, collate soy relevant medical advice into an actionable recommendation on how much tofu to consume.

Let’s get started!

If you think I’m missing important considerations, I would love to hear them in the comments.

If you would like to learn more about tofu or support any way, let me know here.

Thanks for your time—I look forward to working with more of you on popularizing Chinese tofus in the States!

SUMMARY: Ways you can support

  • Founding

    • Co-founding a tofu import company.

    • Founding a tofu CPG (packaged food) brand. (I can share a long list of product ideas and connect you with tofu suppliers.)

    • Opening a ghost restaurant.

    • Organizing a destigmatize soy SWAT team.

    • Writing a bestselling book on Chinese tofu cooking.

  • Funding me

    • $10-20k to commission an R&D chef to develop scalable tofu use cases.

    • $1000 to send tofu boxes to 10 food influencers.

    • $13k to cover my cost of living for 4 months to pursue tofu work full time. (I’m applying to the EA Animal Welfare Fund but am unsure whether my grant will go through.)

    • Angel investment in a tofu import company, once we validate a market, hash out a business model, and have a better sense of budget needs.

  • Promoting

    • This post to your networks.

    • Tofu projects/​products via social and other media.

  • Connecting me with

    • Potential co-founders

    • Chefs, restaurateurs, influencers, food writers, culinary schools, etc.

    • Researchers

  • Researching

    • Case studies—how have other foods become crazy popular?

    • Risk assessment—how can we minimize the risk of cultural appropriation (and its perception)?

    • Literature review—can we lower the isoflavone content of soybeans or tofu?

    • Soy health—collating relevant medical advice into an actionable recommendation on how much tofu to consume (for an MD or something with medical authority.)

  • Advising

    • Around business/​startups

    • Around food industry

    • Around cultural appropriation-relevant risk, branding, or PR.

  1. ^

    I’m working with a small team of chefs on a book to develop new uses for 5 tofus within western cooking. See https://​​www.brokencuisine.com for more info.

  2. ^

    This report suggests that US tofu market sales are around ~$500M. If the average supermarket tofu block is $1.50, this implies that less than 1 block per American is sold each year. https://​​www.grandviewresearch.com/​​industry-analysis/​​tofu-market.

    This market size is different from that in my previous post, which estimated that the average Californian ate 1-3 blocks tofu/​month. Perhaps this latest number is an underestimate, but I doubt it’s off by more than an order of magnitude.

  3. ^

    2020 US per capita meat consumption=115 lbs chicken. https://​​farmdocdaily.illinois.edu/​​2021/​​05/​​an-overview-of-meat-consumption-in-the-united-states.html

    The average broiler chicken liveweight is 6.5 lbs https://​​www.nationalchickencouncil.org/​​about-the-industry/​​statistics/​​u-s-broiler-performance/​​

    65% chicken live weight is edible. https://​​yourmeatguide.com/​​whole-chicken-portions-meat-yield/​​

    (115lbs ) /​ (6.5lbs/​bird * 65%) = 27.2 birds eaten per person /​ per year.

    If 80% of chicken is eaten at lunch and dinner, and tofu displaced 50% of the meat eaten at a given meal, this would mean .015 chickens displaced per meal. Multiplied by the 333 million blocks of tofu eaten per year, that would mean nearly 5 million chickens/​year.

    (27.2birds*80%)/​(365*2)*(333,000,000*50%)=4,900,000 chickens.

    A 50% meat substitution rate would also displace other meat consumption beyond chicken. The impact of reduced beef and pork consumption is likely dwarfed by chicken, as they’re much larger animals. Chicken substitution may itself be dwarfed by fish and seafood substitutions. While meat demand may not immediately lead to lower supply (like if less chicken consumption led to lower demand, which reduced the price, which led to increased consumption...), it should go down over the longer run.

  4. ^

    I’m writing a book on this subject—see https://​​www.brokencuisine.com. The reasons are somewhat convoluted, but if you shoot me a message and your email I can send you a full explanation from our book.

  5. ^

    While some plant-based meat products have taken off, these ingredients are still, overwhelmingly, perceived by chefs and food snobs to be less desirable than meat. This is because most still don’t taste as good, those that taste as good in one use (like burgers) don’t taste as good in other uses, and the products are highly processed. While processing is not inherently bad in my mind, chefs’ favorite ingredients are almost always farm-to-table, natural, artisan, or in-season ingredients, meaning that plant-based meats have an inherent disadvantage.

    In contrast, Chinese tofus aren’t meat analogs; they have unique properties, allowing them to do unique things that meat cannot. They are also natural, healthy, and can have artisan, cultural stories.

    These ingredients, I believe, can turn chefs from our biggest antagonists – the biggest stalwarts to plant-based progress – to our evangelists – literally using these ingredients to create new culinary experiences, serving them to their customers, and beginning to shift America’s gastronomic culture away from meat.

    Granted, this is just a hypothesis—it could be wrong. We should have a better sense once more restaurants begin trialing these ingredients.

  6. ^

    What chefs enjoy cooking and diners love eating.

  7. ^

    Before tofu: chefs dislike plant protein options → restaurant menus are mostly meat.

    After tofu: chefs fall in love with a plant protein → start designing menus around these ingredients (not as substitutes for meat, just as their own thing) → tofu crowds out meat options, resulting in less meat being ordered → consumers get used to eating more plants and normalize different habits.

    The impact comes from crowding out and changing norms, rather than directly replacing meat dishes one-by-one.

    In contrast, to completely replace the meat on a mid-sized restaurant menu, a chef might need 10 great, high fidelity meat/​dairy/​egg analogs. We simply don’t have those yet, and probably won’t for several decades (https://​​forum.effectivealtruism.org/​​posts/​​4uYebcr5G2jqxuXG3/​​when-can-i-eat-meat-again). And most chefs don’t want highly processed plant-based meats, even if they taste as good.

  8. ^

    Most Americans care about animal suffering (https://​​news.gallup.com/​​poll/​​183275/​​say-animals-rights-people.aspx), yet the vast majority eat meat. The chief reason is taste.

  9. ^

    Some people have asked why I focus on tofu, rather than tempeh, lentils, or anther plant-based protein. The reason is that Chinese tofus are far more culinarily diverse than these others. Of the 20+ types, each has unique strengths that can do things that the others cannot. The difference, for example, between a cheesy and a dried tofu is huge—they’re entirely different ingredients.

    (And they’re also delicious, affordable, sustainable, and healthy.)

  10. ^

    Besides a few food bloggers, there aren’t many people promoting Chinese tofus in the States. An even smaller portion are focused on non-Chinese or non-Asian audiences.

  11. ^

    In its simplest form, tofu is bean curd. Think of this like soymilk cheese. Drop something sour or minerally into fresh, hot soymilk and its proteins will coagulate, or bind together, into soft, pillowy curds. These curds can be left as is – to coalesce into silken tofu – or spooned into a mold and pressed into soft tofu, firm tofu, pressed tofu, or tofu sheets. In China, firm and pressed tofu are often smoked, salted, dehydrated, or fermented – which transforms them into several other distinct varieties.

    Not all tofu is bean curd, however. Some varieties are made from soy protein, not whole beans (see spongy tofu). Others are biproducts from making bean curd (see yuba).

    While most tofus are made from soybeans, there are also non-soy varieties. These are generally made from rice, almonds, peanuts, chickpeas, oat chestnuts, hemp, sesame, egg, pig’s blood, or milk. Don’t let their name fool you, though – these ingredients taste, feel, and cook up completely differently from the originals.

    See https://​​georgestiffman.medium.com/​​a-complete-list-of-chinas-27-tofus-7274f764d24 for a long list of different varieties.

  12. ^

    This tofu is not really used outside of Shanghai/​Zhejiang/​Jiangsu cooking; there are maybe 10 Shanghai restaurants in greater LA, and given that roughly 10% of Chinese Americans live in greater LA (https://​​en.wikipedia.org/​​wiki/​​Chinese_Americans; https://​​en.wikipedia.org/​​wiki/​​List_of_U.S._cities_with_significant_Chinese-American_populations), I’d guess that there are less than 100 Shanghai restaurants in the U.S.

    Less than 20% of Shanghai restaurants in LA use this tofu. Those that do might use 10 logs/​week (it’s not popular.) That would imply that restaurants use roughly 10,000 logs per year. If retail demand were similar, that would imply a total yearly demand of just 20,000 logs. Even if retail demand were a lot higher, it still wouldn’t be much.)

    Import-to-wholesale markups are roughly 20%, implying gross revenues of roughly $.50/​log, or $10,000.

  13. ^

    Spongy tofu, fermented tofu, stinky tofu, and tofu sheets all seem similarly uncommon.

    Pressed tofu is more popular, as it’s used across regional Chinese cuisines—maybe there’s a 10-50x bigger market? This is the main type of Chinese tofu that is produced locally.

  14. ^

    One 20′ shipping container should fit 10 pallets of Shanghai tofu, or roughly 20,000 logs.

  15. ^

    LD-3 are the smallest air freight containers with refrigeration (reefer) options that I could find. They can fit roughly 1 pallet, or 1⁄10 of a 20′ container. https://​​targettransport.com/​​wp-content/​​uploads/​​2018/​​09/​​Air-Freight-Container-Specifications.pdf

  16. ^

    The Freightos.com shipping calculator quoted me ~$3000 for a 180kg pallet of 11 cases x 30 packs of suji → $10/​pack, or $5/​log or block-worth. This article has similarly high rates: https://​​huntersourcing.com/​​shipping-from-china-to-us/​​ .

  17. ^

    Talking to restaurants has been surprisingly straightforward. I’ve literally just walked into restaurants during the slow time (or when they’re closed in the afternoon preparing for the dinner rush), asked if their manager was around, and described that I’m thinking of starting a tofu import business and trying to gauge interest. Then we’ll have a brief conversation, I’ll offer to bring samples, and things happen.

    I only just started restaurant outreach so am unsure whether many of these warm leads will become partners.

  18. ^

    Whether it’s something about the dish—flavor, appearance, name, preparation -, restaurant clientele, overall menu composition, etc.

  19. ^

    It’s probably easiest to promote tofu via food service/​restaurants first, then retail/​grocery stores. This is because consumer education is harder for home cooks than restaurants; you can better control the user experience in food service so increase the likelihood of enjoyment; and food service demand seems easier to initially scale. Moreover, a book would need more than 30 recipes (ideally 100+), meaning it would require more upfront investment.

  20. ^

    Promoting tofu “as an ingredient” that chefs can build around could give it legs—resulting in dozens of interesting use cases.

    Promoting it as a product (like a frozen meal) will only ever have one use.

  21. ^

    Letting food bloggers loose scares me. What if they decide we need to press the other tofus?? Or even worse, marinate them!? These techniques have stuck around for a long, long time—even though they don’t work for their stated goals.

  22. ^

    Thinking about The Art of Fermentation, The Food Lab for science-informed cooking, Modernist Cuisine, etc.

  23. ^

    To my knowledge, much less than 5% of alt protein seed stage companies have later grown and won over mainstream meat-eating audiences.

    Moreover, even the top companies, like Impossible and Beyond, have hardly shifted animal suffering. An 888lb average carcass weight for cows (https://​​www.beefmagazine.com/​​beef/​​cattle-slaughter-and-carcass-weights-2021) means that one cow could make over 3,500 quarter-pound burgers. Beyond Meat is projecting ~$400M in revenue in 2022. (https://​​www.reuters.com/​​business/​​retail-consumer/​​beyond-meat-slashes-revenue-forecast-cut-200-jobs-2022-10-14/​​) Assuming that this came entirely from burgers, each burger was directly replacing an animal meat burger, and one burger sold for $2 (for easy math—I think the number is closer to 3 or 4), that would mean they sold 200M burgers, or the equivalent of not even 60,000 cows.

    The impact would be higher for a chicken, shrimp, or fish product, but most companies haven’t approached 1⁄10 of Beyond’s valuation, and most don’t appeal to meat eaters.

    Granted, these numbers are ever shifting. Maybe a new breakout company will emerge. Or maybe Beyond or Impossible will grow another 10x.

    Are 5% odds of producing a bestseller too optimistic? It’s hard to say—plant-based eating is trending up, cultural foods are taking off, this topic has incredible narratives around US-China dialogue/​peacebuilding, much of producing a bestseller is just PR, many folks in the veg and EA communities have written bestsellers and could support such a project...

  24. ^

    Cookbook often take 1-2 years of part-time effort from a solid team, familiar with the cuisine and book publishing. Research-intensive ones might take much longer, but fortunately there are millions of tofu recipes on the Chinese internet that can be referenced and cited.

  25. ^

    Two years ago, a small LA-based virtual restaurant space might cost $2-3k/​month with just a few months required down-payment. Hiring a quality chef might require a bit more. Getting enough tofu could be a challenge if someone can’t start up an import business and the existing importers can’t guarantee supply.

  26. ^

    4.9 million chickens saved each year because of conventional tofu (see note 2) * 1.0 (growth in tofu market) * 10% likelihood *1/​2year (acceleration) = 246,000 chickens saved.

  27. ^

    4.9 million chickens saved each year because of conventional tofu (see note 2) * 1.0 (growth in tofu market) * (15-10% likelihood) = 246,000 chickens saved each year.

  28. ^

    4.9 million chicken saved each year because of conventional tofu (see note 2) * (2-1.0) (growth in tofu market) * (10% likelihood) = 492,000 years of chicken suffering saved each year.

  29. ^

    More rigorous estimates might model growth situations based on use cases, then sum up the weighted probabilities of each use case panning out.

    Take one example use case: Shanghai tofu used as soup dumplings.

    Say Shanghai tofu soup dumpling balls were a great add-in for chicken noodle soup that allowed restaurants to reduce the amount of chicken per serving by 50%, saving money while maintaining the deliciousness of the soup. Say there were 100,000 US restaurants that served 5 servings of chicken noodle soup per day (this is a totally made up number), 5% of them were in California (which would allow for easy tofu distribution), and 5% were eventually attainable as customers. This would imply a “serviceable obtainable market” for this one use case of Shanghai tofu of ~460,000 bowls of soup per year. If one original bowl of soup used 1⁄10 of a chicken, and one adjusted bowl used 1⁄20, then the amount of savable chickens per year would be ~23,000.

    Say there was a 10% chance of achieving that entire market, a 20% chance of achieving 50%, a 30% chance of achieving 20%, and on… Then we could add up those numbers and create a more precise impact estimate.

    (This is a pretty unimpactful use case, as not much chicken is used. Culinary research could definitely uncover better ones.)

    For each use case, there would be a new potential addressable market. Potential impact estimates could roughly add up.

    Research could shift the likelihood percentages, which use cases we end up focusing on (i.e., market size), or the speed at which we come up with them.

  30. ^

    Human studies have looked at soy intake on diverse hormonal indicators (testosterone, fertility, etc.) and found that consuming 60-240 milligrams of isoflavones per day has zero downsides. (https://​​examine.com/​​articles/​​is-soy-good-or-bad/​​)

    For reference, 240 milligrams of isoflavones is roughly two 14 ounce blocks of firm tofu. That’s over 60 grams of soy protein – in one day! (https://​​www.ars.usda.gov/​​ARSUserFiles/​​80400525/​​Data/​​isoflav/​​Isoflav_R2-1.pdf − 100g of Azumaya Firm Tofu has 31 mg of isoflavones, which is high compared to other comparable tofus. Thus, one 14 oz block would have roughly 123 mg isoflavones. Cooking seems to reduce this amount.)

  31. ^

    See section on “PREVALENCE DATA”: https://​​journals.lww.com/​​nutritiontodayonline/​​Fulltext/​​2020/​​01000/​​Recent_Surveys_on_Food_Allergy_Prevalence.6.aspx

  32. ^

    Chickens, the most efficient livestock animal, only convert 11% of the feed they eat into calories. That means to get 110 calories of chicken meat, farmers need to first grow 1000 calories worth of corn and soy feed. Pigs and cows are even less efficient. https://​​awellfedworld.org/​​feed-ratios/​​

  33. ^

    The US government Congressional Research Center has modeled reforestation opportunities and finds that they could significantly reverse climate change. https://​​sgp.fas.org/​​crs/​​misc/​​R40562.pdf

  34. ^

    I have always assumed pea protein to be more expensive than soy but didn’t find good comparisons in a 3 minute search. If anyone has knowledge about this, feel free to chime in.

  35. ^

    https://​​pubmed.ncbi.nlm.nih.gov/​​15137811/​​

  36. ^

    https://​​pubmed.ncbi.nlm.nih.gov/​​27211649/​​

  37. ^

    https://​​pubmed.ncbi.nlm.nih.gov/​​9848521

George Stiffman20 Oct 2022 16:46 UTC
284 points
37 comments18 min readEA link