Maybe this is obvious to those smarter than I, but I'm beginning to have a sinking sense that direct deaths by SARS-COV2 will be a shockingly small minority of total deaths caused by the pandemic over the next generation. We're going to see this event loud and clear in every economic and demographic chart for the next half-century, aren't we?
1. I hate how everything has to have a financialised solution:
2. I read articles like this and wonder, will the world ditch meat, or will will the richer countries continue to eat ridiculous amounts of the stuff, using up all the land that could grow food for a plant based diet? Something will have ti give at some point.
To achieve this, the report suggests “mobilizing international financial support” and implementing tools such as “fertilizer contract swaps” to keep farmer costs manageable.
"The 70 percent of solar energy the Earth absorbs per year equals roughly 3.85 million exajoules. In other words, the amount of solar energy hitting the earth in one hour is more than enough to power the world for one year." 
"The total solar power hitting Earth is about 173,000 terawatts, or 1.73×10^17 joules per second. That’s roughly equivalent to the energy of 41 Megatons of TNT exploding… every second. It’s hard for us to comprehend how much energy a joule is (or even what energy is in the first place). But energy can be converted into mass, and we do understand what mass is." 
Thus, equatorial and semi-equatorial regions should be energy magnats of the world and be able to get nitrogen, phosphorus, water, etc for their food from thin air and sand. Yet, the relevant world has still to feed the majority of these regions, including the solar tech etc.
First issue “fertilizer” is not one thing but a class of things produced in very different ways. The article doesn’t even mention which things are going to have supply problems and why.
To complicate matters Russia currently is using all available train cars and thus has no compacity to increase rail traffic to ship fertilizer.
That is absolutely 100% not the solution to anything for probably a dozen very good reasons. The impact of this kind of "recycling" is well studied and the negatives outweigh the positives.
I mean, I get it. Every programmer salivates at the idea of writing a super efficient hand crafted assembly routine that blows the doors off of what's on the market right now, and that's good.
But I can't help but think... If I wanted to drastically change people's behavior, it can be done easily.
Step one: Cause a supply chain issue, creating artificial scarcity.
Step two: Lecture everyone on how they need to be less wasteful, caution them about how painful life will be if "we aren't all in this together".
Step three: Find every day Joe's and make an example of them in the public sphere for selfishly squandering precious resources. The public will gladly be your enforcers, thinking that if they all try hard enough the artificially created problem will go away.
This technique always results in artificially created human suffering, and waste. The old joke about economic systems and cows was something to the tune of, "you have two cows. The communist milks your cows, and pours the milk down the drain".
It's fascinating to read what Gareth Jones wrote about tractors in the Soviet Union -- how meddlesome people who didn't know how to farm / engineer inserted themselves into the process to tell everyone how to do their jobs, and royally screwed everything up.
Edit: Link for the curious. Day thirty.
> "According to an average of 13 global databases from 10 data sources, in 2010, 161 teragrams of nitrogen were applied to agricultural crops, but only 73 teragrams of nitrogen made it to the harvested crop. A total of 86 teragrams of nitrogen was wasted, perhaps ending up in the water, air, or soil. The new research was published in the journal Nature Food in July."
Large-area applications by mechanized systems seem to be part of the problem, but that's also necessary to escape the subsitence agriculture trap, i.e. with such systems, it's not necessary for half or more of the human population to be working in the fields to grow food, it's more like 1 in 50 or 1 in 100.
The most promising solution might be AI + robots. If a robot could crawl up and down fields inspecting individual plants for nutrient status and applying small amounts of fertilizer as needed (also weeding and checking for pest infestations), it could cut fertilizer use in half while maintaining the same level of production - and perhaps eliminate the need for most herbicides and pesticides.