CO2 and job impacts related to fish consumption by households in Portugal

CO2 impacts from fish consumption in Portugal in 2011 using EIOA & Exiobase 3.0

Figure 1: From: https://pixabay.com/en/fish-market-seafood-fish-428058/

Introduction

The Environmental Input-Output Analysis (EIOA) course offers student the possibility for a deep-dive into this upcoming tool in the field of IE. Firstly introduced in the 70s by Nobel Laureate Wassily Leontief, this technique has been widely used within the field of IE to pursue some interesting analysis on the environmental impacts of economic activity.

I am a big fan of this "complex"and sometimes "abstract" technique because I like to focus on macro interventions in terms of environmental policy analysis. EIOA has shaped my perspective of global CO2 emissions when reading for example Davis & Caldeira (2010) analysis on how global trade shapes CO2 emission pathways, Hertwich & Peters (2009) on the carbon footprint of nations, about the structural composition of CO2 emissions in China Minx et al (2011) or even on the analysis of the CO2 upstream impacts coming from a service based economy like in the US (Suh, 2006). Regardless of the article(s), I think that EIOA is really useful at giving great insight into global supply chain problems when it comes to environmental issues arising from economic activity and in the end life as humans in a society. Why I find it interesting is because it can give straightforward answers to complex environmental policy making decisions. For example, despite the fact that developed countries continuously report a less carbon intensive economy in terms of CO2/GDP, the fact is that we just outsource our emissions to China for example when we import CO2 intensive activity (goods and etc like steel production, mining or the likes) from overseas. This is just one of the pivotal ideas behind carrying out EIOA analysis.

In the mind of beginners in the field of IE, we always try to reason in terms of LCA (the impact of a given product/function) and we neglect the powers of macro thinking. In the sense that, aggregation sometimes is enough to answer a lot of environmental related questions. Doing an LCA on all disaggregated "things" like all the nth products composing industry X would just be too cumbersome and sometimes not really relevant for the question that we're trying to answer (Hendrickson, 2006). Therefore using directly EIOA methods to assess the impacts of industry X may be a better starting point for a lot of environmental policy questions like what are the problems with some manufacturing sectors, why the meat industry is so intensive in terms of US beef versus Chinese chicken, etc, only LCA can give very precise and detailed analysis on the impacts of 1 kg of each product. I don't mind sticking with EIOA because myself I like more to think the world in global terms. I think only thinking global can solve a lot of societal challenges despite each nation's jurisdictions. If you want to learn more about the differences, check this blog entry by one of the leading experts in IE, prof Pauliuk.

I do plan to take the LCA course and become a "certified" LCA expert by prof. Guinee who is a reference in the field. Maybe my opinion will change next semester.

My assignment

Now coming back to my assignment and why I did it. Well I come from a fishing town in the south of Portugal called Quarteira. Fish there is big business and in the past employed a large percentage of the population (would be nice to have reference here - I don't have so just believe me). Over the years we keep on hearing about the EU's restrictive sea catch policies on certain species and how this affects locals. In a nutshell the complex problem can be reduced to the fact that imposing quotas for locals means that they don't have boats large enough to go farther into open ocean and capture fish there like other rich fisherman have (quoting a friend of my mother, Manel). Irony aside, this would be a worrisome problem if there would be no other sources of income in the town/region. Fortunately our airport keeps on pouring tourists at ballooning rates which leads to precarious jobs in the Horeca industry - fisherman destitute of their means of subsistence can always find a job elsewhere. Happened to my friend Bruno.

Figure 2: What some would argue a beautiful view of an overly-constructed patch of land by the sea side. Quoting myself.

Seeing the rise of fish from aquaculture (mainly Greece/Spain) for sale in supermarkets and even in fish markets, I always wondered about the whole picture of what was happening. Meaning that as a lot of sea catch is being replaced by aquaculture fish since sea catch can be 3x more expensive in a lot of cases. I thought for my assignment it would be nice to look at:

1. The impacts in terms of CO2 emissions if more and more fish consumed by Portuguese families would come from aquaculture, mainly from Norway and/or Spain?
2. What would be the impacts in terms of job loss in Portugal and job addition in Norway and Spain?

To answer both questions, Exiobase 3.0 is used because without going into much detail about what it is and why to use it, it's like changing the screen of your broken iPhone without following a YouTube tutorial. Or if you never did that, imagine refurbishing a jet engine without an assembly manual and all parts scattered around. Jokes aside, here's what it is:

"EXIOBASE is a global, detailed Multi-regional Environmentally Extended Supply and Use / Input Output (MR EE SUT/IOT) database. It was developed by harmonizing and detailing SUT for a large number of countries, estimating emissions and resource extractions by industry, linking the country EE SUT via trade to an MR EE SUT, and producing an MR EE IOT from this. The international input-output table that can be used for the analysis of the environmental impacts associated with the final consumption of product groups."

By using something called the Leontief inverse or the total requirement matrix we can model to the nth relationship the supply chain of a given product. For example to make 1 € of potatoes in Turkey, machinery from Germany is needed which in turn requires steel from china which in turn requires iron from somewhere or scrap. This is very useful when we want to account for all the supply-chain related emissions of economic systems.

What I did

Since the aim of this blog is not to share theory rather results and fun (not) facts about IE, I will upload the code that I used in Python for reproduction and leave matrix multiplication where it belongs. The exercise I did is a very simple and straightforward analysis that is light years behind what researchers in the field of IE are doing like our Professor João F.P. Rodrigues. Still the results can answer to a certain extent the research question and that is nice, right?!

In a nutshell what I did is this:

1. See how much € Portuguese households spent in this 2 product categories of our economy: "Fish products" and "Fish and other fishing products; services incidental of fishing". The first, "Fish products" gathers what we do with fish when we transform it. The second is about fish itself sold fresh or frozen.
2. Isolate how much Portugal consumed from Portugal itself (1114 m€) meaning that Portuguese households consumed 1114 m€ of fish from Portuguese output. The total consumption accounting for imports was 2180 m€
3. Allocate the slice of domestic spending to importing either from Norway or from Spain while reducing the amount of internal consumption. 
4. Check how CO2 emissions and jobs change when we switch from domestic consumption to imports.

Results

CO2 emissions 

Total CO2 emissions in the Exiobase region resulting from the allocation of those 557m€ from the Portuguese households either into Norway or Spain result in a decrease of 77 kt of CO2 when importing from Norway or in an increase of 114 kt of CO2 if coming from Spain.

The Portuguese fishing industry in this case emits 74% less CO2 than the Spanish one while when compared to the Norwegian, the Portuguese ones emits 1.6 times more CO2 per m€ of economic activity. This highlights the fact that Norway is "greener" in terms of CO2 emission related to the fish industry therefore increasing activity in Norway will contribute to global CO2 emission reduction.

Next, what is the change in CO2 emissions from the fishing industry in the 3 countries? We can see that when allocating demand to imports from Norway or Spain, total Portuguese CO2 emissions from the fishing industry drop as reported in the figure below.

 

In terms of % change, the difference is shown below:

In absolute terms, the CO2 emission drops in the fishing sectors are -35% in Portugal, ± 7% increase in Norway and ±13% increase in Spain. 

Jobs

In terms of thousand job changes. As expected Portugal would "suffer" a lot if more fish related goods would be imported.

this is because Portugal has a very intensive job market in terms of numbers of jobs needed per m€ of fish stuff. This is 2.2 times more than Spain and 15 times more than Norway! Which leads me to conclude that the Portuguese fishing industry is much more labor intensive than the Spanish or Norwegian. This obvious in the following figure where the % job gain/loss is shown for the three scenarios:

Portugal will lose around 38% jobs if those 557 m€ would come from Norway or from Spain. On the contrary Norway would gain ± 10% more jobs and Spain ± 12% more jobs.

Conclusion

• Actually not much besides what is kind off logic: Norway has a lot of fish production from aquaculture which quite less intensive in terms of CO2 emissions. But as pointed out Buchspies et al. (2011), Norway mostly relies on green electricity and the majority of the environmental impacts from aquaculture are actually in other impact categories like ecotoxicity. 
• Portugal and Spain have an inefficient fishing industry that emits substantial amounts of CO2 when compared to others
• In terms of jobs the same applies, lots of job losses and family drama in Portugal if the population starts consuming more fish from imported aquaculture.
• Maybe I should have done the analysis for GWP of CO2eq and not only of CO2 --> sorry!

Here's the GitHub code if you feel like having some "fun"

Sources: 

1. Davis, S. J., & Caldeira, K. (2010). Consumption-based accounting of CO2 emissions. Proceedings of the National Academy of Sciences, 107(12), 5687-5692.
2. Hertwich, E. G., & Peters, G. P. (2009). Carbon footprint of nations: A global, trade-linked analysis. Environmental science & technology, 43(16), 6414-6420.
3. Minx, J. C., Baiocchi, G., Peters, G. P., Weber, C. L., Guan, D., & Hubacek, K. (2011). A “carbonizing dragon”: China’s fast growing CO2 emissions revisited. Environmental science & technology, 45(21), 9144-9153. 
4. Sangwon Suh (2006). Are services better for climate change? Environmental science & technology 40(21). 6555-6560
5. Hendrickson, C. T., Lave, L. B., Matthews, H. S.  (2006).  Environmental Life Cycle Assessment of Goods and Services:  An Input-Output Approach.  Resources for the Future Press. 
6. Buchspies, B., Tölle, S. J., & Jungbluth, N. (2011). Life cycle assessment of high-sea fish and salmon aquaculture. ESU-services Ltd., fair consulting in sustainability, Uster, Switzerland.