Sunday, October 15, 2017

Endogenous Growth Models and Stagnation in Agricultural Innovation

See also: Ideas and Research Productivity in Agriculture 

From: Big Ideas Are Getting Harder to Find - Stanford Business Insights:

"big ideas are getting harder and harder to find, and innovations have become increasingly massive and costly endeavors, according to new research from economists at the Stanford Institute for Economic Policy Research. As a result, tremendous continual increases in research and development will be needed to sustain even today’s low rate of economic growth." 

One of the key equations in the paper relates the number of researchers working in an area to the number of new ideas: research productivity = (A*/A)/S(t) = # new ideas / # researchers One particular area they looked at was with productivity in agriculture, measuring ideas as crop yields.

 “For instance, to measure productivity in agriculture, the study’s co-authors used crop yields of corn, soybeans, wheat and cotton and compared them against research expenditures directed at improving yields, including cross-breeding, bioengineering, crop protection and maintenance....On average, research productivity in agriculture fell by about 4% to 6% per year, the study found....Research productivity is simply the ratio of average yield growth divided by the number of researchers." 

They were careful to tease out research related to non-yield related traits. However, is 'yield' always the best way to measure productivity? Maybe it is a good proxy, but it looks like their data really starts in the 60's....about the time when corn seed was being transitioned away from double cross to single cross hybrids, which were more uniform and higher yielding. In terms of having a large marginal impact on yield, that seems like it could be a tough act to follow, even 40+ years later (its not like we will rediscover hybrid vigor or heterosis) no matter how many PhD's we throw at increasing yield. Other developments in crop improvement may in fact be inflating the denominator in the equation above. The convergence of big data and genomics has opened the door to numerous potential lines of research related to crop improvement, that although may only marginally impact yield provide other significant technological benefits to growers, consumers, and the environment. This would be important, even if yield were constant.

Advances in big data and genomics may have resulted in many more possible needles to search for in a much greater number of haystacks than before. Each crop has a different genome and a seemingly infinite number of biochemical pathways that could be of interest. Then there is the micro-biome and who knows what is next. We just need way more researchers than before.

 I'm not sure how much this applies in other fields, and in some ways this could be making the authors' point....look at all the lines and directions of research and scientists pursuing them vs. actual improvements in crop yields. But again, even if yield were constant, for every bushel of corn/soy produced today, how much have we improved with regard to CO2 reduction, energy use, erosion/leaching, reduction in toxic chemical exposure (thanks largely to roundup and Bt traits), and biodiversity (in terms of non-pest targets)?

Yield just seems to be one outcome among many of importance to creating sustainable development in the agriculture space (although it is important enough to sustainability that we certainly would not want to revert back to older technologies, genetics, and chemistries as some have argued). It would be interesting to see this modeled with traits or variety patents or total hybrids as a measure of new ideas.

References:

Are Ideas Getting Harder to Find? By Nicholas A. Bloom, Charles I. Jones, John Van Reenen, Michael Webb. September 2017Working Paper No. 3592

Modern Corn and Soybean Production. Rober G. Hoeft, Emerson D. Nafziger, Richard R. Johnson, and Samuel R. Aldrich. MCSP. First Edition. 2000.

Environmental impacts of genetically modified (GM) crop use 1996–2015: Impacts on pesticide use and carbon emissions. Graham Brookes & Peter Barfoot. GM Crops & Food Vol. 8 Iss.2, 2017 Genetically Engineered Crops: Has Adoption Reduced Pesticide Use?

Agricultural Outlook ERS/USDA Aug 2000 Greenhouse gas mitigation by agricultural intensification Jennifer A. Burneya,Steven J. Davisc, and David B. Lobella.PNAS June 29, 2010 vol. 107 no. 26 12052-12057

Friday, October 13, 2017

Ideas and Research Productivity in Agriculture

From:

Big Ideas Are Getting Harder to Find - Stanford Business Insights

https://www.gsb.stanford.edu/insights/big-ideas-are-getting-harder-find?utm_source=TWITTER&utm_medium=Social&utm_campaign=Insights&Date=20171013&linkId=43474574



“For instance, to measure productivity in agriculture, the study’s co-authors used crop yields of corn, soybeans, wheat and cotton and compared them against research expenditures directed at improving yields, including cross-breeding, bioengineering, crop protection and maintenance.

The average yields across all four crops roughly doubled between 1960 and 2015. But to achieve those gains, the amount of research expended during that period rose “tremendously” – anywhere from a threefold to a more-than-25-fold increase, depending on the crop and specific research measure.

On average, research productivity in agriculture fell by about 4% to 6% per year, the study found.”

Diving into the NBER paper referenced in the article above:

"The faster rising number corresponds to research targeted only at so-called biological efficiency. This includes cross-breeding (hybridization) and genetic modification directed at increasing yields, both directly and indirectly via improving insect resistance, herbicide tolerance, and efficency of nutrient uptake, for example…The slower-growing number additionally includes research on crop protection and maintenance, which includes the development of herbicides and pesticides. ….It is immediately evident from Figure 6 that research productivity has fallen sharply for agricultural yields: yield growth is relatively stable or even declining, while the effective research that has driven this yield growth has risen tremendously. Research productivity is simply the ratio of average yield growth divided by the number of researchers."

The authors seemed really careful to tease out differences in R&D devoted to yield enhancing technology vs other innovations in crop improvement. However, is 'yield' always the best way to measure productivity? What do we mean by productive?

I had a course in grad school related to economic growth (I recall lots of time devoted to the Solow model and endogenous technological chcange) but I am not a growth or developmental economist so I could be getting off track here.

But, for instance, the lines of research they are attributing to yield (insect resistance, herbicide tolerance, and efficency of nutrient uptake) have lots of other positive effects related to reductions in CO2, reduced erosion and groundwater pollution, improved biodiversity, etc. Even if yield were constant, for every bushel of corn/soy produced today, how much have we improved with regard to CO2, erosion/leaching, reduction in toxic chemical exposure (thanks largely to roundup and Bt traits), and biodiversity (in terms of non-pest targets)?

I suppose I understand the infatuation with yield and why it matters, but it is challenging to think that agriculture is experiencing stagnation in growth given all that is going on in that industry (like CRISPR and the convergence of big data and genomics).

Friday, September 01, 2017

Voter Irrationality and Systematic Bias: Applications in Food and Biotechnology

In the Myth of the Rational Voter: Why Democracies Choose Bad Policies, Bryan Kaplan discusses issues related to the median voter theorem and systematic biases by voters.

One interesting concept he discusses is the miracle of aggregation. According to the miracle of aggregation democracies can make decisions as if all were well informed. If we assume that less informed voters make random mistakes,  errors will cancel and the votes that matter will be the informed ones. The well informed median voter determines the outcome.

This all breaks down if the most informed voters make systematic mistakes. In that case the median preference becomes biased away from the optimal policy. But why would well informed voters make systematic mistakes?

Sometimes our values and views are part of who we are. Believing certain things gives people higher levels of utility. They let preferences drive beliefs over evidence. To entertain information or evidence to the contrary would upset preferences and lower utility. To quote Caplan:

"letting emotions or ideology corrupt thinking is an easy way to satisfy such preferences"

He also quotes Lebon:

"the masses have never thirsted after the truth, they turn aside from evidence that is not to their taste...whoever can supply them with illusions is easily their master; whoever attempts to destroy their illusions is always their victim"

This idea of preferences driving beliefs explains a lot. For instance, the election of demagogues. There are clear benefits to be reaped in customizing political platforms and media content that feeds into the preferences of these different segments of the population. The media capitalizes on that at the expense of actually informing the electorate. So do politicians and pundits.

This also may explain the explosion of growth in organic, natural, hormone free and other niche food markets.  Or the popular support for GMO labeling initiatives despite the science behind both safety and environmental benefits of biotechnology.

All of these are cases where acceptance of scientific evidence should potentially change opinions and behavior as it relates to food and agriculture. However to change those opinions and choices would be to drastically upset the preferences of a number of consumers. This makes it hard for those in agriculture and science communication trying to help the public navigate the complex world of modern agriculture. It also makes it hard for companies, wanting to do the right thing, to make a stand for science (i.e. by not going down the non-GMO/hormone/gluten free negative labeling route).

For instance, what if a t-shirt manufacturer wanted to promote the use of Bt cotton in their products on the basis of a reduction in use of toxic pesticides and improved insect biodiversity? Or what if a food company wanted to promote their dairy products for having a lower carbon footprint due to rBST? Taking this position would likely upset the illusions and preferences held dearly by many consumers. Noone wants to become 'their victim' to borrow from Lebon. Just ask Monsanto or BPI, the company behind finely textured beef. (however ABC eventually paid a price for feeding the masses the pink slime 'illusion'). In response, we don't see these kinds of promotions, and to the contrary we actually see companies removing these technologies from their product lines (and advertising the fact!).

Due to systematic bias in relation to food and technology, the median of voters' preference distribution will be biased toward more restrictive regulations than is scientifically appropriate. This will influence the types of products we see on the shelves and the potential for healthier and more environmentally sustainable solutions to challenging worldwide problems.

References:

The Myth of the Rational Voter: Why Democracies Choose Bad Policies, Bryan Kaplan

The Crowd: A Study of the Popular Mind.  Gustave Le Bon.

A Meta-Analysis of Effects of Bt Cotton and Maize on Nontarget Invertebrates.Michelle Marvier, Chanel McCreedy, James Regetz, Peter Kareiva Science 8 June 2007: Vol. 316. no. 5830, pp. 1475 – 1477

Areawide Suppression of European Corn Borer with Bt Maize Reaps Savings to Non-Bt Maize Growers. Science 8 October 2010:Vol. 330. no. 6001, pp. 222 - 225 DOI: 10.1126/science.1190242W. D. Hutchison,1,* E. C. Burkness,1 P. D. Mitchell,2 R. D. Moon,1 T. W. Leslie,3 S. J. Fleischer,4 M. Abrahamson,5 K. L. Hamilton,6 K. L. Steffey,7, M. E. Gray,7 R. L. Hellmich,8 L. V. Kaster,9 T. E. Hunt,10 R. J. Wright,11 K. Pecinovsky,12 T. L. Rabaey,13 B. R. Flood,14 E. S. Raun15

The environmental impact of recombinant bovine somatotropin (rbST) use in dairy production Judith L. Capper,* Euridice Castañeda-Gutiérrez,*† Roger A. Cady,‡ and Dale E. Bauman* Proc Natl Acad Sci U S A. 2008 July 15; 105(28): 9668–9673

The environmental impact of dairy production: 1944 compared with 2007. Journal of Animal Science,Capper, J. L., Cady, R. A., Bauman, D. E. 2009; 87 (6): 2160 DOI: 10.2527/jas.2009-1781

Environmental impacts of genetically modified (GM) crop use 1996–2015: Impacts on pesticide use and carbon emissions
Graham Brookes & Peter Barfoot
GM Crops & Food Vol. 8 , Iss. 2,2017

Batista R, Saibo N, Lourenço T, Oliveira MM. Microarray analyses reveal that
plant mutagenesis may induce more transcriptomic changes than transgene
insertion. Proc Natl Acad Sci U S A. 2008 Mar 4;105(9):3640-5. doi:
10.1073/pnas.0707881105. PubMed PMID: 18303117; PubMed Central PMCID: PMC2265136

Baudo MM, Lyons R, Powers S, Pastori GM, Edwards KJ, Holdsworth MJ, Shewry PR. (2006). Transgenesis has less impact on the transcriptome of wheat grain than conventional breeding. Plant Biotechnol J. 2006 Jul;4(4):369-80

Sunday, August 27, 2017

Single Peaked Preferences and the Median Voter Theorem

In the last entry on voting paradoxes, I mentioned that things are different if preferences are single peaked.

Let’s look at another scenario. Again, consider a set of policies A,B,C following a sequence from  (A) less to more extreme (C) - maybe tax rates or some level of spending- with (B) being the intermediate policy. Suppose voters rank policies in order of preference/utility as follows:

Voter X:  ABC
Voter Y : CBA
Voter Z : BCA

In this case, no matter what order is undertaken, B always ends up being the law that is enacted. These preferences are single peaked. Each individual has a most preferred choice along the A to C spectrum. If you move away from that choice (in the A-B-C spectrum) they prefer the other choices less.  In the previous example, voter Y did not have single peaked preferences and that is what caused the cycling or order dependent outcomes. With single peaked preferences there is a new problem. With single peaked preferences, the median point of the preference distribution will elicit the most votes. Only those laws or candidates with a centrist twist will get the majority of the votes. Only those voters with centrist views will be happy, and it makes it very difficult for candidates to be elected if they want to bring about major reforms. This phenomenon is referred to as the ‘median voter theorem.’

Both voter cycling (when preferences are not single peaked) and the median voter theorem can have negative implications for majority rule policy adoption. Voters are either governed by irrational ever changing majorities, or they are subjugated by entrenched majorities whose views are maintained by the status quo of median preferences.

References:

Lemieux, Pierre, The Public Choice Revolution. Regulation, Vol. 27, No. 3, pp. 22-29, Fall 2004. Available at SSRN: https://ssrn.com/abstract=604046

Majority Rule and Vote Cycling with Non-Single Peaked Preferences

Is majority rule the best way to represent voters preferences for a given set of policies?

Let’s look at a particular voting scenario to illustrate this. Consider a set of policies A,B,C following a sequence from  (A) less to more extreme (C) - maybe tax rates or some level of spending- with (B) being the intermediate policy. Suppose voters have the following preferences:


VOTER X: A >B >C  'single peaked '

VOTER Y: C>A>B

VOTER Z: B>C >A  'single peaked'

Both voters X and Z have single peaked preferences. As we move away from their optimal choice the strength of their preferences or utility decreases. However, voter Y does not exhibit single peaked preferences. They prefer the extreme policy C most, but their next preferred policy is in the direction of the other extreme A. They prefer the intermediate policy B least.

If the voters were voting on these issues, voter X would prefer law A over law B and law B over law C. In shorthand – A >B >C. To summarize all of the choices of the voters we see that 2/3 of the voters have preference A >B, 2/3 of the voters have preference B > C, but when voting A vs. C, 2/3 have preference C >A.

See if you follow the application of this. If we vote on policies in a pairwise fashion and have two elections and the first is made between policy B and C, then B will win (2/3 of the voters have preference B > C). If this is followed by a second election A vs. B (Because C was eliminated in the first election) then A will be the law that ultimately passes by majority rule.


Now if the order is changed, in which the first election is between A and B, A will win (because 2/3 of the voters rank A > B). Then in the second election when A goes against C, C will be the law that passes by majority rule (again because 2/3 of the voters have preference C >A).

So when voting on these policies, the process becomes arbitrary. The outcome depends on the order of the vote, so a cycling of choices ensues. According to public choice economist Gordon Tullock, any outcome can be obtained in majority voting by at least one voting method. This indicates that majorities can be irrational and dangerous unless preferences are all single peaked. 

Reference:

Lemieux, Pierre, The Public Choice Revolution. Regulation, Vol. 27, No. 3, pp. 22-29, Fall 2004. Available at SSRN: https://ssrn.com/abstract=604046

Thursday, August 24, 2017

Risk, Uncertainty, Speculation and Granger Causality

What is the difference between 'risk' and 'uncertainty'?

"Uncertainty refers to outcomes that we cannot foresee, or whose probabilities that we cannot estimate. In other words, uncertainty is a way of characterizing what we don't know about the distribution of the random variables themselves...Risk can be quantified, priced,and traded. It can even be hedged with large pools of assets." - Froeb et al 2014.Managerial Economics: A problem solving approach. 3rd edition.

This distinction was first made by economist Frank Knight.

Insurance of all kinds and commodity futures markets are examples of financial products based on quantifiable risks. Those involved in these markets, often speculators, have an important role to play. Speculators and futures and options markets make it possible to allocate resources across time, essentially from periods of abundance to periods of scarcity smoothing consumption and alleviating risks.

This is described well in The Economic Way of Thinking:

"Speculators coordinate market exchanges through time...that both inform people and provide them with the opportunity to allocate their risks...Speculators accept the risk at a mutually agreed upon price that hedgers seek to avoid."

Elaine Kub, author of Mastering the Grain Markets puts it beautifully in her book:

"All life on earth depends on agriculture, how well we distribute agriculture's products-how well we trade grain-determines how Earth's population gains access to its most fundamental needs."

However, speculators and funds involved in commodities trading often come under heavy scrutiny perhaps without understanding the important role they play in actually increasing food security as Nevil Speer explains below:

“Reining in speculators seems politically expedient.  But we live in complex times.  Throwing darts becomes perilous when policy makers begin to advocate (and worse yet, actually believe) that speculators should be removed  from ag / food markets.   Such a move would dismantle futures markets.  Imagine what the world might look like a without market liquidity, price discovery and risk mitigation; not to mention the inability to establish pricing plans, attract new capital investment and stimulate innovation across the food business.   The absence of those influences, facilitated by futures markets, would ultimately lead to less food production, availability and security – NOT the other way around.  Taking speculators out of the mix would be devastating.”  Dr. Nevil Speer, No Speculators? No thanks!, Drovers Cattle Network Agsight, March 2011

So what impact have speculators had on markets? One way is to look at the impact of index funds on commodity prices and volatility. Economist Scott Irwin at the University of Illinois has looked at this in depth. Below are three examples of studies (although not a complete review of the literature) looking at these impacts based on granger causality tests and other methods:

The Impact of Index Funds in
Commodity Futures Markets:
A Systems Approach
DWIGHT R. SANDERS AND SCOTT H. IRWIN
The Journal of Alternative Investments
Summer 2011, Vol. 14, No. 1: pp. 40-49

"The system of Granger-style causality tests fails to reject the null hypothesis that that trader positions do not lead market returns. Hence, there is no evidence of a linkage between index trader positions in commodity futures markets and price levels."

Irwin, S. H. and D. R. Sanders (2010), “The Impact of Index and Swap Funds on Commodity Futures Markets: Preliminary Results”, OECD Food, Agriculture and Fisheries Working Papers, No. 27, OECD Publishing. doi: 10.1787/5kmd40wl1t5f-en

“There is no statistically significant relationship indicating that changes in index and swap fund positions have increased market volatility......at this time, the weight of evidence clearly suggests that increased index fund activity in 2006-08 did not cause a bubble in commodity futures prices.”

Index Trading and Agricultural Commodity Prices:
A Panel Granger Causality Analysis
Gunther Capelle-Blancard and Dramane Coulibaly
CEPII, WP No 2011 – 28
No 2011 – 28
December

"Our results show that, in agricultural futures markets, there is no evidence of a causality relationship from index funds to futures prices. This result holds for the period 2006-2010, but also for the sub-periods 2006-2008 and 2008-2010. These findings imply that index-based trading has not been an important driver in the substantial increase in commodities prices. Changes in commodity prices may instead reflect fundamental supply and demand factors."


References:

Explained: Knightian uncertainty
The economic crisis has revived an old philosophical idea about risk and uncertainty. But what is it, exactly?
Peter Dizikes, MIT News Office
June 2, 2010

Foeb, Mcann, Ward, and Shor. Managerial Economics: A problem solving approach. 3rd edition. 2008.

The Economic Way of Thinking. Heyne, Boettke, and Prychitko. 10th Edition. 2002.

Mastering the Grain Markets: How profits are really made. Elaine Kub.

See also: 

Fat Tails, Kurtosis, and Risk

Fat Tails, the Precautionary Principle and GMOs

Friday, July 07, 2017

Stawman Arguments Against Statements Related to GMO Safety

Previously I discussed how the World Health Organization, the American Medical Association, and the National Academy of Sciences have all issued statements regarding the safety of foods derived from genetically engineered crops.

In addition I discussed how critics have questioned these statements. One set of assertions supports the invocation of the precautionary principle.

I have written before about issues related to using the precautionary principle with respect to genetically modified vs conventional food crops.

In this post I would like to specifically discuss the assertion that  "There are no epidemiological studies investigating potential effects of GM food consumption on human health."

To those unfamiliar with modern crop science and genetics, that could sound like a very condemning statement. But that begs the question, have there been epidemiological studies investigating the potential effects of conventionally and mutagenically improved crops on human health?

Its also a true statement that there are no epidemiological studies investigating the relative safety of using the stairs vs. elevators vs. escalators vs. leaping out the top floor window with regard to human health. (although I am sure actuaries have assessed property/casualty probabilities associated with similar kinds of risks related to building design, we don't have people losing sleep over lack of publication in this area)

These last examples might seem extreme and unrelated, but they illustrate the point that for some things, conducting an expensive (and difficult) epidemiological study to assess impacts on human health makes little practical sense. 

What reasoning would make us think this is necessary for genetically modified foods? If we were discussing inclusion of traits known to impact metabolism or hormone levels or some other biological function this might make sense. But the types of crops approved for human consumption don't have traits known to behave this way. Some critics might assert that it is the unknown consequences (changes in DNA, changes in proteins, or metabolism) that we should be worried about. 

However, scientists know that these kinds of genetic disruptions are not any more proliferate with genetically engineered crops than those related to traditional and mutagenic crop improvement that have been consumed and accepted by consumers without question for hundreds (thousands) of years or more in some cases and decades in others. They are substantially equivalent in this regard.

It turns out that the statement about the absence of epidemiological studies is really irrelevant when it comes to assessing the risks associated with genetically engineered food consumption. Arguments using epidemiological studies to form a psychological baseline or frame of reference are akin to strawman statements that could raise unnecessary doubts and fears about a technology that actually exhibits characteristics beneficial to human health and the environment.

References:

No scientific consensus on GMO safety. Environmental Sciences Europe. 2015 27:4

Batista R, Saibo N, Lourenço T, Oliveira MM. Microarray analyses reveal that
plant mutagenesis may induce more transcriptomic changes than transgene
insertion. Proc Natl Acad Sci U S A. 2008 Mar 4;105(9):3640-5. doi:
10.1073/pnas.0707881105. PubMed PMID: 18303117; PubMed Central PMCID: PMC2265136

Baudo MM, Lyons R, Powers S, Pastori GM, Edwards KJ, Holdsworth MJ, Shewry PR. (2006). Transgenesis has less impact on the transcriptome of wheat grain than conventional breeding. Plant Biotechnol J. 2006 Jul;4(4):369-80
First citation:

Monday, July 03, 2017

Defining Consensus Regarding the Safety of Genetically Modified Foods

The World Health Organization, the American Medical Association, and the National Academy of Sciences have all issued statements regarding the safety of foods derived from genetically engineered crops. However, in a recent letter critical of the documentary film Food Evolution, the following paper is cited:


No scientific consensus on GMO safety. Environmental Sciences Europe. 2015 27:4
Link

This was the first time I had seen this paper so I spent some time going through it to see what kinds of arguments were being made. Below are a few excerpts and some discussion.

"the scarcity and contradictory nature of the scientific evidence published to date prevents conclusive claims of safety, or of lack of safety, of GMOs. Claims of consensus on the safety of GMOs are not supported by an objective analysis of the refereed literature."
"The health, environment, and agriculture authorities of most nations recognize publicly that no blanket statement about the safety of all GMOs is possible and that they must be assessed on a 'case-by-case' basis."

"There are no epidemiological studies investigating potential effects of GM food consumption on human health"

"an expert panel of the Royal Society of Canada issued a report that was highly critical of the regulatory system for GM foods and crops in that country. The report declared that it is 'scientifically unjustifiable' to presume that GM foods are safe without rigorous scientific testing and that the 'default prediction' for every GM food should be that the introduction of a new gene will cause 'unanticipated changes' in the expression of other genes, the pattern of proteins produced, and/or metabolic activities."

"We support the application of the Precautionary Principle with regard to the release and transboundary movement of GM crops and foods."

 I have not had a chance to check every single reference and citation made. However the general framework sketched out in the paper I am getting is this:


  • there is no absolute or conclusive evidence that genetically engineered foods are safe or unsafe
  • the risks are associated with unintended effects related to gene insertions (i.e. genetic disruptions)
  •  invocation of the precautionary principle is used to obviate the statements often cited by the World Health Organization, the American Medical Association, and the National Academy of Sciences

This leads me to ask, can we make a blanket statement about the safety of all conventionally modified or organic foods that utilize plant breeding and mutagenesis? Have there been epidemiological studies investigating the effects of these methods on human health?

Suddenly this thinking brings up a question I have addressed before: why would we invoke the precautionary principle in the case of  food from genetically engineered crops and not for conventionally and mutagenically improved crops? 

From the literature:

“We found that the improvement of a plant variety through the acquisition of a new desired trait, using either mutagenesis or transgenesis, may cause stress and thus lead to an altered expression of untargeted genes. In all of the cases studied, the observed alteration was more extensive in mutagenized than in transgenic plants” - (Batista, et al; 2008)

With greater disruptions, critics might favor increased regulatory scrutiny. However, we do not have a framework in place for mutagenically improved crop varieties that have been safely used for decades and approved by the organic food industry and accepted by consumers, nor do we have anything like this for conventionally bred crops. If an argument for the precautionary principle holds for genetically engineered crops on this basis, then it should also hold for all types of crop improvement.

Therefore it seems tenuous to make a scientific risk based justification for special treatment of genetically engineered crops without further evidence. When many refer to a consensus on the safety of genetically engineered foods, this is what I have in mind.

Policies related to genetically engineered foods leveraging the precautionary principle could lead to increased risk of doing more harm than good to human health and the environment if policies prevent or delay adoption of traits that could decrease use of toxic pesticides, or reduce carbon emissions and improve soil conservation as some biotech traits have been shown to do in the literature.

See also:

Fat Tails, The Precautionary Principle, and GMOs
Comments on Rules for Gene Editing Technology
Organic Activists Realize Hypocrisy in Opposition to Gene Editing Technology

References:

Batista R, Saibo N, Lourenço T, Oliveira MM. Microarray analyses reveal that
plant mutagenesis may induce more transcriptomic changes than transgene
insertion. Proc Natl Acad Sci U S A. 2008 Mar 4;105(9):3640-5. doi:
10.1073/pnas.0707881105. PubMed PMID: 18303117; PubMed Central PMCID: PMC2265136

Baudo MM, Lyons R, Powers S, Pastori GM, Edwards KJ, Holdsworth MJ, Shewry PR. (2006). Transgenesis has less impact on the transcriptome of wheat grain than conventional breeding. Plant Biotechnol J. 2006 Jul;4(4):369-80