Less global inequality can improve climate outcomes

By Narasimha Rao, Project Leader of the Decent Living Energy (DLE) Project, IIASA Energy Program

Is there a conflict between reducing global income inequality and combating climate change? This seems like an odd question, given that these challenges have a lot in common. Raising the living standard of the poor for example, makes them resilient to climate impacts; less inequality can mean more political mobilization to establish climate policies; and changes in social norms away from material accumulation can reduce inequality and emissions. Academics have however been curious about the following phenomenon: In many countries, a dollar spent at higher income levels is less energy intensive than at lower income levels (known as “income elasticity of energy”). That is, rich people – although they consume much more in total – spend additional income on services or can afford energy-efficient goods, while the new middle class buy energy-intensive goods, like appliances and cars.

Many imagine China as a template for this type of fast growth. If globally significant, this effect would imply that growth that is more equitable would also be more emissions-intensive, and that we would have to pay particular attention to ensuring that climate policies reach the rising middle class in developing countries. While several studies have examined this phenomenon in specific countries, no one has examined its global significance. We set out to do that.

Energy intensity (MJ per $) lower in a high-growth, low inequality world (green line, Gini=0.29) compared to a low-growth, high inequality world (blue line, Gini=0.45). Gini reflects between-country inequality only.

Our analysis suggests that the energy-increasing effect of lowering inequality is more of a distraction than a concern. We compared scenarios of equitable and inequitable income growth, both within and between countries, assuming the most extreme manifestation of the income elasticity. Within any country, given the slow pace at which inequality typically evolves even with the most extreme known income elasticity and reduction in country inequality, greenhouse gas emissions would increase by less than 8% over a couple of decades. However, when one considers a more equitable distribution of growth between countries, global emissions growth may decrease when compared to growth that occurs in industrialized countries. This is because poorer countries have more potential for technological advancements that reduce the energy intensity of growth than richer countries do. That is, more income growth in poorer countries provides more opportunity for efficiency improvements that influence the emissions of very large populations. Furthermore, China is a poor model for poor countries at large, many of which have relatively low energy intensities, even today.

Climate stabilization at the level aspired to by the Paris Climate Agreement requires that we (i.e. the world) decarbonize to zero annual emissions around 2050, which means that even developing countries have to make aggressive strides towards integrating climate goals into development. Nevertheless, there is no sufficient basis for considering that equitable growth, and by implication the poor’s energy intensity, is part of the problem. To the contrary, the potential for co-benefits from equitable growth for climate change are enormous, but unfortunately under-explored, particularly in quantitative studies. Research should focus on quantifying the role of changing social norms – less consumerism, political mobilization, and other social changes that are typically associated with lower inequality – on reducing greenhouse gases. ­

Reference:

Rao, ND, Min J. Less global inequality can improve climate outcomes. Wiley Interdisciplinary Reviews: Climate Change. 2018;e513. https://doi.org/10.1002/wcc.513

Note: This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

Using Twitter data for demographic research

By Dilek Yildiz, Wittgenstein Center for Demography and Global Human Capital (IIASA, VID/ÖAW and WU), Vienna Institute of Demography, Austrian Academy of Sciences, International Institute for Applied Systems Analysis

Social media offers a promising source of data for social science research that could provide insights into attitudes, behavior, social linkages and interactions between individuals. As of the third quarter of 2017, Twitter alone had on average 330 million active users per month. The magnitude and the richness of this data attract social scientists working in many different fields with topics studied ranging from extracting quantitative measures such as migration and unemployment, to more qualitative work such as looking at the footprint of second demographic transition (i.e., the shift from high to low fertility) and gender revolution. Although, the use of social media data for scientific research has increased rapidly in recent years, several questions remain unanswered. In a recent publication with Jo Munson, Agnese Vitali and Ramine Tinati from the University of Southampton, and Jennifer Holland from Erasmus University, Rotterdam, we investigated to what extent findings obtained with social media data are generalizable to broader populations, and what constitutes best practice for estimating demographic information from Twitter data.

A key issue when using this data source is that a sample selected from a social media platform differs from a sample used in standard statistical analysis. Usually, a sample is randomly selected according to a survey design so that information gathered from this sample can be used to make inferences about a general population (e.g., people living in Austria). However, despite the huge number of users, the information gathered from Twitter and the estimates produced are subject to bias due to its non-random, non-representative nature. Consistent with previous research conducted in the United States, we found that Twitter users are more likely than the general population to be young and male, and that Twitter penetration is highest in urban areas. In addition, the demographic characteristics of users, such as age and gender, are not always readily available. Consequently, despite its potential, deriving the demographic characteristics of social media users and dealing with the non-random, non-representative populations from which they are drawn represent challenges for social scientists.

Although previous research has explored methods for conducting demographic research using non-representative internet data, few studies mention or account for the bias and measurement error inherent in social media data. To fill this gap, we investigated best practice for estimating demographic information from Twitter users, and then attempted to reduce selection bias by calibrating the non-representative sample of Twitter users with a more reliable source.

Exemplar of CrowdFlower task © Jo Munson.

We gathered information from 979,992 geo-located Tweets sent by 22,356 unique users in South-East England and estimated their demographic characteristics using the crowd-sourcing platform CrowdFlower and the image-recognition software Face++. Our results show that CrowdFlower estimates age more accurately than Face++, while both tools are highly reliable for estimating the sex of Twitter users.

To evaluate and reduce the selection bias, we ran a series of models and calibrated the non-representative sample of Twitter users with mid-year population estimates for South-East England from the UK Office of National Statistics. We then corrected the bias in age-, sex-, and location-specific population counts. This bias correction exercise shows promise for unbiased inference when using social media data and can be used to further reduce selection bias by including other sociodemographic variables of social media users such as ethnicity.  By extending the modeling framework slightly to include an additional variable, which is only available through social media data, it is also possible to make unbiased inferences for broader populations by, for example, extracting the variable of interest from Tweets via text mining. Lastly, our methodology lends itself for use in the calculation of sample weights for Twitter users or Tweets. This means that a Twitter sample can be treated as an individual-level dataset for micro-level analysis (e.g., for measuring associations between variables obtained from Twitter data).

Reference:

Yildiz, D., Munson, J., Vitali, A., Tinati, R. and Holland, J.A. (2017). Using Twitter data for demographic research, Demographic Research, 37 (46): 1477-1514. doi: 10.4054/DemRes.2017.37.46

Note: This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

Cornelius Hirsch: Digging into foreign investment in agriculture

By Parul Tewari, IIASA Science Communication Fellow 2017

Two things are distinctly noticeable when you meet Cornelius Hirsch—a cheerful smile that rarely leaves his face and the spark in his eyes as he talks about issues close to his heart. The range is quite broad though—from politics and economics to electronic music.

Cornelius Hirsch

After finishing high school, Hirsch decided to travel and explore the world. This paid off quite well. It was during his travels, encompassing Hong Kong, New Zealand, and California, that Hirsch started taking a keen interest in economic and political systems. This sparked his curiosity and helped him decide that he wanted to take up economics for higher studies. Therefore, after completing his masters in agricultural economics, Hirsch applied for a position as a research associate at the Austrian Institute of Economic Research and enrolled in the PhD-program of the Vienna University of Economics and Business to study trade, globalization, and its impact on rural areas. Currently, he is looking at subsidies and tariffs for farmers and the agricultural sector at a global scale.

As part of the 2017 Young Scientists Summer Program at IIASA, Hirsch is digging a little deeper to analyze how foreign direct investments (FDI) in agricultural land operate. “Since 2000, the number of foreign land acquisitions have been growing—governmental or private players buy a lot of land in different countries to produce crops. I was interested in knowing why there are so many of these hotspots in the world— sub-Saharan Africa, Papua New Guinea, Indonesia—why are people investing in these areas?,” says Hirsch.

Farming in one of the large agricultural areas in Indonesia ©CIFOR I Flickr

Increased food demand from a growing world population is leading to an increased rate of investment in agriculture in regions with large stretches of fertile land. That these regions are largely rain-fed make them even more attractive for investors as they save the cost of expensive irrigation services. In fact, Hirsch argues that “the term land-grabbing is misleading. It should actually be water-grabbing as water is the foremost deciding factor—even more important than simply land abundance.”

Some researchers have found an interesting contrast between FDI in traditional sectors, such as manufacturing, and the ones in agricultural land. While investors in the former look for stable institutions and good governmental efficiency, FDI in land deals seems to target regions with less stable institutions. This positive relationship between corruption and FDI is completely counterintuitive. Hirsch says that one reason could be that “sometimes weaker institutions are easier to get through when it comes to such vast amount of lands. A lot of times these deals and contracts are oral and have no written proof—the contracts are not transparent anyway.”

For example in South Sudan, the land and soil conditions seem to be so good that investors aren’t deterred despite conflicts due to corrupt practices or inefficient government agencies.

One of the indigenous communities in Madagascar, a place which is vulnerable to land acquisitions © IamNotUnique I Flickr

One area that often goes unnoticed is the violation of land rights of indigenous communities. If a government body decides to sell land or give out production licenses to investors for leasing the land without consulting the actual community, it is only much later that the affected community finds out that their land has been given away. Left with no land and hence no source of livelihood, these communities are forced to migrate to urban areas.

A strain of concern enters his voice as Hirsch talks about the impact. “Land as big as two times the area of Ecuador has been sold off in the past—but it accounts for a tiny percentage of the global production area.” With rising incomes and greater consumption of meat, a lot of land is used to produce animal feed crops. “This is a very inefficient way of using land,” he says.

During the summer program at IIASA, Hirsch is generating data that will help him look at these deals in detail and analyze the main factors that are taken into consideration before finalizing a land deal. At the moment he is only able to give an overview of land-grabbing at the global level. With more data on the location of the deals he can look at the factors that influence these decisions in the first place such as the proximity between the two countries involved in agricultural investments and the size of their economies.

While there is always huge media coverage when a scandal about these land acquisitions comes out in the open, Hirsch seems determined to dig deeper and uncover the dynamics involved.

About the researcher
Cornelius Hirsch is a research associate at the Austrian Institute of Economics and Research (WIFO). At IIASA he is working under the supervision of Tamas Krisztin and Linda See in the Ecosystems Services and Management Program (ESM).

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

Do smokers know what they are doing to their life expectancy?

By Valeria Bordone, University of Munich Department of Sociology and IIASA World Population Program

Everyone, consciously or unconsciously, formulates in their own mind a subjective survival probability– i.e., an estimate of how long they are going to live. This will affect decisions in different spheres of later life: retirement, investments, and healthy behaviors. Moreover, previous research has found that subjective survival probability is a good predictor of mortality. In fact, on average, people somehow know better than standard health measures the effect that their characteristics and their behavior have on life expectancy. It is however plausible not only to expect differences within the population in terms of survival, but also in the ability to predict their own survival.

(cc) roujo | Flickr

In a recent publication with Bruno Arpino from the University Pompeu Fabra and Sergei Scherbov from the Wittgenstein Centre (IIASA, VID/ÖAW, WU)., we presented for the first time joint analyses of the effect of smoking behavior and education on subjective survival probabilities and on the ability of survey respondents to predict their real survival, using longitudinal data on people aged 50-89 years old in the USA drawn from the Health and Retirement Study.

We found that, consistent with real mortality, smokers report the lowest subjective survival probabilities. Similarly, less educated people report lower subjective survival probabilities than higher education people. This is in line with the well-known positive correlation between education and life expectancy. However, despite being aware of their lower life expectancy as compared to non-smokers and past smokers, people currently smoking at the time of the survey tended to overestimate their survival probabilities. This holds especially for less educated people.

This graph shows the probability of correctly estimating the own survival probabilities with 95% confidence intervals, by smoking behavior and educational attainment. ©Arpino B, Bordone V, & Scherbov S (2017)

Our study suggests that in fact, education also plays an important role in shaping people’s ability to estimate their own survival probability. Whether or not they smoke, we found that more highly educated people are more likely to correctly predict their survival probabilities.

In view of the high proportion of the American population that consists of current or past smokers, a percentage that reached 77% in some male cohorts, our findings emphasize the need to disseminate more information about risks of smoking, specifically targeting people with less education.

By showing that smoking and education play together in determining how well people can assess the own survival potential, this study extends our understanding of the variability of subjective survival probabilities within a population. The fact that sub-groups within the population differently incorporate the effects of smoking into their assessment of survival probabilities may have important consequences for example on when people exit the labor market or whether they buy a life insurance, as individuals are likely to base their decisions also on their longevity expectations.

Policymakers can therefore draw some relevant conclusions from our study to design policies concerned with health and survivorship in later life. Despite the various anti-smoking campaigns and smoking restrictions, smokers may not be fully aware of the risks of smoking. In particular, educational groups seem to be differently exposed to the information that is disseminated to the public. Our study suggests that there is a need to target such information to less educated people, who are the most likely to underestimate the risks of smoking. Providing information on how survival probabilities vary by smoking behavior may not only reduce smoking but it may also increase individuals’ ability to assess their own survival.

(cc) Quinn Dombrowski | Flickr

Reference
Arpino B, Bordone V, & Scherbov S (2017). Smoking, Education and the Ability to Predict Own Survival Probabilities: An Observational Study on US Data. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-17-012 [pure.iiasa.ac.at/14692]

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.

What would an oil spill mean for the Arctic?

By Parul Tewari, IIASA Science Communication Fellow 2017

As climate change warms up the planet, it is the Arctic where the effects are most pronounced. According to scientific reports, the Arctic is warming twice as fast in comparison to the rest of the world. That in itself is a cause for concern. However, as the region increasingly becomes ice-free in summer, making shipping and other activities possible, another threat looms large. That of an oil spill.

©AllanHokins I Flickr

While it can never be good news, an oil spill in the Arctic could be particularly dangerous because of its sensitive ecosystem and harsh climatic conditions, which make a cleanup next to impossible. With an increase in maritime traffic and an interest in the untapped petroleum reserves of the Arctic, the likelihood of an oil spill increases significantly.

Maisa Nevalainen, as part of the 2017 Young Scientists Summer Program (YSSP), is working to assess the extent of the risk posed by oil spills in the Arctic marine areas.

“That the Arctic is perhaps the last place on the planet which hasn’t yet been destroyed or changed drastically due to human activity, should be reason enough to tread with utmost caution,” says Nevalainen

Although the controversial 1989 Exxon Valdez spill in Prince William Sound was quite close to the Arctic Circle, so far no major spills have occurred in the region. However, that also means that there is no data and little to no understanding of the uncertainties related to such accidents in the region.

For instance, one of the significant impacts of an oil spill would be on the varied marine species living in the region, likely with consequences carrying far in to the future. Because of the cold and ice, oil decomposes very slowly in the region, so an accident involving oil spill would mean that the oil could remain in the ice for decades to come.

Thick-billed Murre come together to breed in Svalbard, Norway. Nevalainen’s study so far suggests that birds are most likely to die of an oil spill as compared to other animals. © AllanHopkins I Flickr

Yet, researchers don’t know how vulnerable Arctic species would be to a spill, and which species would be affected more than others. Nevalainen, as part of her study at IIASA will come up with an index-based approach for estimating the vulnerability (an animal’s probability of coming into contact with oil) and sensitivity (probability of dying because of oiling) of key Arctic functional groups of similar species in the face of an oil spill.

“The way a species uses ice will affect what will happen to them if an oil spill were to happen,” says Nevalainen. Moreover, oil tends to concentrate in the openings in ice and this is where many species like to live, she adds.

During the summer season, some islands in the region become breeding grounds for birds and other marine species both from within the Arctic and those that travel thousands of miles from other parts of the world. If these species or their young are exposed to an oil spill, then it could not only result in large-scale deaths but also affect the reproductive capabilities of those that survive. This could translate in to a sizeable impact on the world population of the affected species. Polar bears, for example, have, on an average two cubs every three years. This is a very low fertility rate – so, even if one polar bear is killed, the loss can be significant for the total population. Fish on the other hand are very efficient and lay eggs year round. Even if all their eggs at a particular time were destroyed, it would most likely not affect their overall population. However, if their breeding ground is destroyed then it can have a major impact on the total population depending on their ability and willingness to relocate to a new area to lay eggs, explains Nevalainen.

Due to lack of sufficient data on the number of species in the region as well as that on migratory population, it is difficult to predict future scenarios in case of an accident, she adds. “Depending on the extent of the spill and the ecosystem in the nearing areas, a spill can lead to anything from an unfortunate incident to a terrible disaster,” says Nevalainen.

©katiekk I Shutterstock

It might even affect the food chain, at a local or global level. “If oil sinks to the seafloor, some species run the risk of dying or migrating due to destroyed habitat – an example being walruses as they merely dive to get food from the sea floor,” adds Nevalainen. As the walrus is a key species in the food web, this has a high probability of upsetting the food chain.

When the final results of her study come through, Nevalainen aims to compare different regions of the Arctic and the probability of damage in these areas, as well as potential solutions to protect the ecosystem. This would include several factors. One of them could be breeding patterns – spring, for instance, is when certain areas need to be cordoned off for shipping activities, as most animals breed during this time.

“At the moment there are no mechanisms to deal with an oil spill in the Arctics. I hope that it never happens. The Arctic ecosystem is very delicate and it won’t take too much to disturb it, and the consequences can be huge, globally,” warns Nevalainen.

About the Researcher

Maisa Nevalainen is a third- year PhD student at the University of Helsinki, Finland. Her main focus is on environmental impacts caused by Arctic oil spills, while her main research interests include marine environment, and environmental impacts of oil spills among others. Nevalainen is working with the Arctic Futures Initiative at IIASA over the summer, with Professor Brian Fath as her supervisor and Mia Landauer and Wei Liu as her co-supervisors.

This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.