Knowledge for better food systems

Animal efficiencies, animal welfare: either/or, or both/and? Some Reflections from an Informal Meeting

Friday, January 16, 2015

Elin Röös and Tara Garnett

Late December last year Tara Garnett, researcher on food systems and climate change and coordinator of the FCRN, initiated a meeting that brought together Marian Dawkins, Prof in Animal Behaviour at Oxford University, Jude Capper, researcher and livestock sustainability consultant who has worked mainly in the US, and Elin Röös, Swedish LCA researcher and the initiator of the Swedish Meat Guide,  for an informal discussion on the subject of sustainable intensification of agriculture and what that entails for the animal welfare of farm animals.

During the course of this meeting, a number of assumptions were challenged, diverse ethical and philosophical questions raised, and some areas that need further exploration and research identified.  The purpose of this article is to share our reflections with the FCRN community and to seek out your thoughts and ideas about what sort of research and activity is needed next.

We started the meeting by inviting Jude to give us a brief summary of some of her research.  In a comparative study of the environmental impact of US beef production in 1977 and 2007, she found that present day beef production requires considerably fewer resources. For example, the beef industry in 2007 only required 88 % of the water and 67 % of the land to produce the equivalent amount of meat as in 1977. She argued that contrary to consumer perceptions, the efficiency gains in the US livestock sector have actually decreased the environmental impact of meat production per kg of meat.   

A similar study by Cederberg et al. (2009) on Swedish livestock production came to the same conclusion; as efficiency increases environmental impact per kg of product go down. This is also well established by numerous LCA studies on livestock products and is the driving rationale underpinning industry efforts to reduce the environmental impacts of production. However, there is (at least) one complicating factor that was raised in the Swedish study and in several LCA studies, some published quite a while ago (Cederberg et al., 2003). As milk production per cow increases, fewer dairy cows are needed to produce the same amount or milk (and so fewer dairy calves are born), which in turn leads to less beef meat being produced from the milk system. If this loss in beef meat output is substituted for by more suckler beef production – suckler beef meat is associated with higher greenhouse gas emissions per kg of meat than beef from dairy systems - then total emissions from beef production will increase. This illustrates the importance of considering the environmental impacts from the livestock sector at a wider system level – a point upon which we all agreed.

To incorporate these considerations into industry decision making is of course very challenging, since systems are becoming increasingly specialized.  There is a need for policy action, but what could such policy look like and how can production systems be designed to avoid sub-optimizing? This is definitely an area that deserves more attention.     

So in general more ’efficient’ production systems reduces greenhouse gas emissions, water and land use per kg of product. How do we then handle the trade-off between increased productivity and its impact on animal welfare?  As Elin asked, is there not an inherent link between increased productivity and reduced animal welfare?  

It depends, said Marian, on which systems we’re talking about. This may indeed be the case right now in systems that are already highly productive, such as those we find in high income countries.  In these contexts measures to improve productivity deliver fairly marginal gains, gains that are often at the expense of animal welfare.  But in many developing regions, where animals are affected by diseases and malnutrition, measures to improve feeding, veterinary and housing conditions hold potential to achieve both improvements in welfare and in productivity.

But ultimately, beyond a certain level of productivity gain that trade off will inevitably arise?

Marian replied that this is not necessarily the case -  or rather, that we actually don’t know. Given current approaches to breeding, yes.  But there are several examples from nature where animals grow tremendously rapidly, without any negative effects on their health. For example, birds that breed in the arctic such as the Red Knot (Calidris canutus) and Greater Snow Goose (Chen caerulescens) have adapted to the short summers by having chicks that grow much faster and are independent much earlier than their temperate counterparts. We have not fully explored can be done through genetics and through breeding strategies if health, robustness and wellbeing were really to be prioritised alongside productivity goals.  To date, the focus of breeders has been almost entirely on improving productivity.  We need to investigate whether it’s possible to breed for traits that achieve both improved health and welfare and productivity levels. Maybe we can, maybe we can’t. But we really don’t know.

Jude agreed and also pointed out the huge variability among animals – she mentioned a dairy cow in the US that produced 32 000 kg milk over the course of one lactation and was perfectly healthy. Her view was that a lot can be done with better breeding programs, but the problem is that so far there haven’t really been any incentives to include animal welfare aspects.

So although breeding for productivity has traditionally been at the expense of animal welfare (Oltenacu & Broom, 2010), it may or may not inherently be the case that the two goals are incompatible.  We need more research to look at this further, as a recent paper by Dawkins and Layton argues.

But is environmental sustainability just about ‘efficiency’? A feeding regime might be environmentally ‘efficient’ expressed in terms of output per volume of GHGs emitted, but surely that’s a rather narrow definition of sustainability? What about resilience and the effective use of resources?  Livestock reared on ‘ecological leftovers’- land unsuited to crop production, or crop residues and other forms of by-products – can act as recyclers of resources, providers of soil fertility, managers of landscapes, and can help keep carbon stored in soils.  By contrast while animal systems with a high feed conversion efficiency may generate fewer emissions expressed in terms of kg CO2eq/kg meat or milk, they will be reliant on soy and other grains which, arguably could be consumed more efficiently directly by humans.  What is more, these feed crops may themselves be vulnerable to the impacts of climatic and other forms of environmental change.  Perhaps a more sustainable system is one that is not just efficient but also more resource effective, more adaptable, more able to bounce back from shocks.  Is it possible to create livestock systems capable of utilising a wider range and more variable mix of feedstuffs including waste and by-products - that are also productive ? And how can we establish incentives for this multi-trait breeding? More exploration and research in this area is certainly needed.

As the discussion continued. Tara  pointed out that so far the conversation about animal welfare had all been about its health dimensions.  But isn’t animal welfare about much more than this?

Marian emphasised the need to start with health. There is so much that can be done just with implementing what we know. Considering the scale of the livestock industry, small relative improvements in mortality rates, reductions in lameness and so forth could potentially lead to huge improvements in the quality of life of many animals.

Rapid growth in the livestock sector is leading to the development of large-scale intensive units where proper management may not be in place. Things can go terribly wrong – poor welfare, excessive use of antibiotics and heightened zoonotic disease risks.  But Jude pointed out that it’s too simplistic to equate large-scale production with poor welfare, and small-scale or extensive systems with good welfare.  Very large units can afford to employ several specialists, someone focused on feeding, someone who deals with breeding, someone in charge of veterinary care and so forth, and this expertise can really deliver good welfare.  This specialist input is simply not possible in a smaller farm where maybe one or two people have to run the whole business.  Poorly-managed small and medium-scale operations exist too.  So good management is key in all systems, but at the moment it is certainly not guaranteed.

The conversation then turned to more controversial approaches to increasing productivity, such as the use of the growth promotor, recombinant bovine somatrophin (rBST).  The hormone is banned in the European Union and in many other parts of the world, and for many Europeans its use would be unacceptable, but in the US the hormone is routinely used. Jude has published several papers on hormones and related products and concludes that their use can be a valuable approach to improving efficiency .  She says:

“Although it’s often suggested that using rbST “burns out” cows by improving their production, at the population level we don’t see that the hormone treated dairy cows are culled at a younger age than unsupplemented cows. The replacement rates in the US and EU are quite similar. It’s important to note that rbST isn’t a silver bullet that will increase production at any cost, it will only do so if the animals are fed and cared for correctly.“ Here Tara commented that the high replacement rates are surely a problem that need to be focused on in all systems.

Whatever the merits of large versus small-scale production, the fact is that both coexist and there are welfare issues across systems.  We all generally agreed that health is a priority area that we can all work to promote regardless of our underlying values. Real and measurable improvements can be achieved without getting bogged in endless discussions about what animal welfare is and how it should be defined.  Since improvements in animal health also bring economic gains to the producers, there is also an obvious incentive for producers to get involved.

But Elin wondered whether by focusing on health as a starting point, we might be at risk of ‘locking ourselves’ into systems that are ultimately limited in their capacity to deliver welfare over and above physiological good health.  Perhaps action on animal health needs to be situated more explicitly within wider discussions about what sorts of animal production we actually want and what we mean by good welfare?

These are very difficult questions and different stakeholders will have different views.  We have a lot of ways of measuring 'good welfare' but no agreement on what is the 'best' way of keeping animals. Some people think we should be aiming for universal free-range, for example, whereas other people disagree, putting emphasis on a different set of welfare indicators. If one thinks that 'behaving naturally' is part of welfare, one may end up favouring a different system than if one thinks that health and having what they want is more important. What is clear though, is that animal welfare considerations deserve more attention than it has received so far, particularly within the GHG  mitigation and sustainable intensification community.  Indeed debates about animal welfare, our relationship with animals - and indeed the ethics of rearing and eating them at all - are themselves just part of a bigger set of much needed discussions about what we really want for the food system and what ultimately we mean by development and human progress.

Finally, what about consumption?  Is the goal ultimately to produce more with less impact – or should the priority be to challenge the “growth is good” agenda?  Demand for meat and milk is growing rapidly across the world.  Is this growth inevitable or should we be seeking to promote alternative consumption trajectories?

Increasingly research is finding that production side gains in efficiency cannot by themselves deliver sufficient reductions in greenhouse gas emissions and that shifts in consumption towards less meat intensive diets can yield important savings.  One might question what the reduced environmental impacts achieved by efficiency gains are worth if increases in meat and dairy consumption cancel out these reductions.  What is more, current trends in eating patterns are putting an unsustainable burden on health services around the world. While meat and animal products can be an important source of nutrients for many, high and growing intakes are associated with a range of chronic diseases.  In the UK, Scandinavia and other parts of Europe there is a growing focus by the scientific and NGO community on the potential health and environmental merits of shifting away from meat-dominant diets. 

What about the US? Jude commented that there is little discussion about consumption within the mainstream in the US.   She felt that it may be possible to convince people to cut back on livestock products in New York and Los Angeles but for most people, the only factor that might change things would be increased meat prices.

On that note, it was time to wrap up. We all recognised that we come at the issues from different perspectives and with different views and values. We agreed on many things and disagreed on a few, but we felt that interdisciplinary open discussions such as this one can be helpful in prompting us question our assumptions, see things from other perspectives, and identify shared questions that may be worth exploring further.  

As a starting point we identified the following areas for potential future research:

  • How are welfare challenges similar and different in different countries, systems and contexts?  Can we identify areas of commonality and cultural, ideological and economic differences?
  • To what extent can productivity, environmental improvements and welfare goals be aligned?  What is the role of animal breeding in this?
  • Should there be more research into more multifunctional systems of production and if so how could industry be persuaded to get involved?
  • What do we know about attitudes to, understandings of and priorities around animal welfare in low income and emerging economies?
  • To what extent do measures to promote environmental efficiency (expressed as volume of output relative to environmental impact) lock us into modes of production and associated consumption that ultimately stimulate growth that negates and even outweighs the benefits of improved efficiency?

We’d be keen to hear the thoughts of the FCRN community on this short article and on the questions raised.  We want to encourage an open discussion - please do get back to us with your comments, with links to relevant research and perhaps with ideas for how we might take forward work in this area. Please submit your comments in the comments box below.


Peter Stevenson's picture
Submitted by Peter Stevenson (not verified) on


The FCRN reflections paper on animal efficiencies and animal welfare argues that industrial livestock production is efficient and decreases the environmental impact of meat production.  Much of the evidence points in the opposite direction.  Industrial livestock production is dependent on feeding human-edible grain to animals.  A recent Chatham House paper describes this process as “staggeringly inefficient”.[i]

Studies, including a UNEP report, show that for every 100 calories that we feed to animals in the form of human-edible crops, we receive on average just 17-30 calories in the form of meat and milk.[ii] [iii]  A 2013 University of Minnesota paper indicates that the efficiency rates may be even lower for some animal products.  It reports that for every 100 calories of grain that we feed to animals, we get only about 40 new calories of milk, 22 calories of eggs, 12 of chicken, 10 of pork, or 3 of beef.[iv] 

A 2013 FAO report points out that the feeding of cereals to livestock could threaten food security by reducing the grain available for human consumption.[v]  Olivier De Schutter, until recently UN Special Rapporteur on the right to food, states that “continuing to feed cereals to growing numbers of livestock will aggravate poverty and environmental degradation”.[vi]  It will aggravate poverty by pushing up cereal prices placing them out of reach for the world’s poor.

Animals’ inefficiency in converting human-edible crops into meat and milk brings other inefficiencies in its train.  More arable land and blue and grey water are generally needed to produce a unit of nutrition from industrially produced meat than from meat derived from animals that are fed little or no human-edible crops.  In considering water use it is important to utilise the distinction between green (rainwater), blue (surface and groundwater) and grey (pollution) water.  The use of green water is generally less important than blue and grey water use (unless farming’s green water consumption competes with other uses and drives them to using blue water).   Mekonnen and Hoekstra concluded that animal products from industrial systems generally consume more blue and grey water than animal products from grazing or mixed systems.[vii]

We similarly need to distinguish between different types of land: pasture which cannot readily be used for other purposes, pasture which can readily be used for other purposes and arable land.  Grazing ruminants on the first kind of pasture is generally efficient as the land could not be used for other forms of food production.  However, using arable land to produce feed crops (other than as part of a traditional rotation) is inefficient; many more people can be fed from a given area of arable land if it is used to produce crops for direct human consumption rather than feed crops.  The Chatham House paper states that only grazing on land unsuitable for crop production or utilizing residues and co-products as feed can be considered an efficient use of land.

The use of arable land to produce feed crops has driven the intensification of crop production.  This has been a disaster for soil quality.  A new UK study reports that the soils resulting from years of industrial agriculture are of poorer quality than those of urban allotments.[viii] 

Turning to animal welfare, I agree with Elin Röös that focusing on health, crucially important as this is, risks locking ourselves into systems that are ultimately limited in their capacity to deliver welfare over and above physiological good health.  We have become very unambitious in our definition of what constitutes good welfare. We need systems that not only prevent disease, injuries and distress but also enable animals to have a positive experience of life.  For this, the ability to carry out natural behaviours is vital as is the provision of sufficient space.  So too is access to fresh air, daylight and the warmth of the sun.  St Basil of Caesaria understood this when sixteen centuries ago he wrote: “May we realize that they live not for us alone, but for themselves and for Thee and that they love the sweetness of life even as we”.  There’s not much scope for sweetness of life in industrial systems.

The FCRN paper suggests that it may be possible to breed animals for greater productivity without adverse impacts on welfare.  This is somewhat speculative while the hard reality is that each year millions of farm animals experience serious health problems as a result of selection for fast growth or high yields.  The European Food Safety Authority has concluded that “long term genetic selection for high milk yield is the major factor causing poor welfare, in particular health problems, in dairy cows”.[ix]   EFSA has also concluded that genetic selection of pigs for rapid growth has led to leg disorders and cardiovascular malfunction.[x]  A UK study into leg disorders in broilers found that, primarily due to high growth rates, 27.6% of the chickens had levels of lameness that are likely to be painful.[xi]   The high productivity of modern laying hens causes osteoporosis which results in a high level of bone fractures.[xii]

To conclude, efficiency and good welfare are compatible with one another; neither, however, is achievable in industrial systems.



[i] Bailey R et al, 2014.  Livestock – Climate Change’s Forgotten Sector.  Chatham House.

[ii] Lundqvist, J., de Fraiture, C. Molden, D., 2008. Saving Water: From Field to Fork – Curbing Losses and Wastage in the Food Chain. SIWI Policy Brief. SIWI.

[iii] Nellemann, C., MacDevette, M., Manders, et al. (2009) The environmental food crisis – The environment’s role in averting future food crises. A UNEP rapid response assessment.  United Nations Environment Programme, GRID-Arendal,

[iv] Cassidy E.M et al, 2013. Redefining agricultural yields: from tonnes to people nourished per hectare.  University of Minnesota. Environ. Res. Lett. 8 (2013) 034015

[v] Gerber, P, 2013. Tackling climate change through livestock – A global assessment of emissions and mitigation

opportunities. Food and Agriculture Organization of the United Nations (FAO), Rome.

[vii] Mekonnen M and Hoekstra A, 2012. A global assessment of the water footprint of farm animal products. Ecosystems.: DOI: 10.1007/s10021-011-9517-8

[viii] Edmondson et al, 2014.  Urban cultivation in allotments maintains soil qualities adversely affected by conventional agriculture.  Journal of Applied Ecology 2014, 51, 880–889

[ix] Scientific Opinion of the Panel on Animal Health and Welfare on a request from European Commission on welfare of dairy cows. The EFSA Journal (2009) 1143, 1-38.

[x] Scientific Opinion of the Panel on Animal Health and Welfare on a request from the Commission on

Animal health and welfare in fattening pigs in relation to housing and husbandry. The EFSA Journal (2007) 564, 1-14

[xi] Knowles, T. G., Kestin, S. C., Haslam, S. M., Brown, S. N., Green, L. E., Butterworth, A., Pope, S. J., Pfeiffer, D. and Nicol, C. J., 2008. Leg disorders in broiler chickens: prevalence, risk factors and prevention. Plos one 3 (2): e1545. doi: 10.1371/journal.pone.0001545.

[xii] Laywell: Welfare implications of changes in production systems for laying hens: Deliverable 7.1

Tara Garnett's picture
Submitted by Tara Garnett on

Dear Peter

Thanks for your comments.  I would just like to comment that this paper reports on the discussions and comments of four people, all of whom hold diverse views - the paper itself does not draw any conclusions at all but rather raises a set of questions for further exploration. 

I should emphasise that while the paper (accurately) observes that life cycle analyses find that intensive system achieve greater efficiencies defined in terms of GHG emissions per unit of animal output it does not thereby conclude that efficiency equates with sustainability.  Indeed the use of the word efficiency is put in inverted commas and just a little later on in the paper it is questioned whether a focus on efficiency equates in any meaningful sense to environmental efficiency. It makes the point that there are other considerations, including resilience, the recycling of nutrients and the use of land unsuited to other purposes to consider. 

As regards issues of water use, arable feed requirements and so forth inherent in intensive systems - you will find that the FCRN has discussed them at length elsewhere (see for example the two 2010 FCRN publications on intensive versus extensive systems and on livestock, feed and food security, as well as the 2009 paper in Environmental Science and Policy - all available in the FCRN publications section of the website). You will also find that I was on the advisory group of the Chatham House project whose publication you cite.

As to the relationship between productivity increases and animal welfare - this was a question that the paper raised as a result of the discussions between myself, Elin, Jude and Marian.  We did not come to any conclusions.  All we did was ask whether the  assumptions that people just like me tend to make - ie. that greater productivities are inherently associated with reduced welfare - really are valid.... or whether further research is needed based on different breeding criteria. 

Finally, you omit to mention the point raised in the paper that a focus on production without  consideration of consumption side strategies (ie. dietary shift) was unlikely to be equal to the challenge of addressing consumption.

I will end by reiterating the point I made right at the start of my response: this is a discussion paper that reflects on the conversations held by four people drawn from different disciplines and with different points of view.  Its intention was to open up questions for further investigation, rather than to close down discussions or to advocate any one particular course of action.

Thanks for taking the time to comment.

Add new comment

Member input

Plain text

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.