Location: Europe

Combating Misuse of Public Funds in COVID-19 Emergency Procurement

Posted on September 28, 2020 | Policy Brief

Image of two medical workers with face shields representing COVID-19 procurement

The Covid-19 pandemic has revealed substantial shortcomings in central governments’ and municipalities’ ability to procure items needed in the fight against Covid-19, and corruption has been rampant partially due to the increased discretion of procurement staff to award contracts. We argue that suspension of ex ante rules safeguarding accountability is essential for disaster relief, but must be compensated for by better ex post monitoring. Such monitoring can be greatly strengthened by increasing transparency of all awarded contracts and providing incentives to whistleblowers to come forward to report fraud and corruption.

Corruption in Covid-19 Procurement

The disastrous Covid-19 pandemic has revealed weaknesses in global supply chains and in national public procurement systems’ ability to secure essential Personal Protective Equipment (PPE), ICU material, and Covid tests. Several countries have been and are experiencing issues like poor quality of procured goods, extremely high prices, scams, and a general inability to source.

Examples of quality under-provision abound. The Spanish government discovered that out of 340,000 tests purchased from a Chinese manufacturer, 60,000 of them did not test accurately for Covid-19 [1], and the Dutch ministry of health issued a recall of 600,000 face masks from a Chinese supplier due to poor quality [2]. Analogous problems were common in the UK [3, 4]. Several countries have also had difficulties to procure at all, for example in terms of their desired number of tests [5, 6], or the reagents used to analyze the tests [7], as well as swabs [8].

Reports on price gouging – selling at extremely high prices – are also widespread. Examples of price gouging and investigations by competition authorities can be found throughout Europe and the US, but also in developing countries like Indonesia, Brazil, Thailand, Kenya, and South Africa (OECD 2020a), and in Ecuador and Paraguay, with corruption as the alleged cause [9].

While many reasons lie behind these procurement failures, several of them are directly traceable to the abuse of the increased discretion granted by emergency procurement rules to urgently source material and bypass time-consuming public procurement processes and legal frameworks. This important and necessary increase in discretion can easily be abused to hand out contracts to friends and/or political allies or to cash bribes.

Again, examples in the press abound. In the UK, a clearly non-urgent contract was awarded without competition to a firm owned by two long term associates of Michael Gove and Dominic Cummings [10]. In Slovenia, a gambling mogul with no public record of healthcare experience appears to have received millions in an emergency contract related to Covid-19 [11]. In Bosnia, a raspberry farm was apparently granted a contract to import 100 ventilators,paying $55,000 for each ventilator, while their price was around $7,000 to $30,000 on the international market in the relevant period [12]. In India, a Mumbai Realtor with no previous healthcare experience got a contract to supply things such as oxygen cylinder and medical beds [13]. The health minister in Bolivia was arrested in May after the country bought 179 ventilators at $27,683 each while it later was revealed that the manufacturers were offering ventilators at approximately half that price [14]. In Bangladesh, Transparency International issued a study suggesting widespread corruption in the country during Covid-19, including the purchase of substandard medical supplies at five to ten times the market price [15].

The Covid-19 crisis has exacerbated an already significant problem: according to Transparency International (2020), up to 25% of all global healthcare procurement spending is lost to corruption.

Historically, Fraud Increases During Emergencies

Disaster related fraud is frequently a problem in the western world as well. In September of 2005, in the aftermath of Hurricane Katrina in the US, the Hurricane Katrina Fraud Task Force was set up to go after frauds related to recovery funds. By August 30th, 2007, the task force had prosecuted 768 individuals for Katrina-related fraud, and additional state and local prosecutions for disaster-related fraud had been brought (DoJ 2007). The National Center for Disaster Fraud was also created within the justice department in the aftermath of several devastating hurricanes in the US, and currently houses over 80 employees.

Organizations and academics warned the public early about the risk of increased corruption in public procurement during the Covid-19 pandemic (Khasiani et al 2020, OECD 2020b). Indeed, emergency procurement and disaster relief has historically been linked to increases in corruption (Leeson and Sobel, 2008), especially where institutions are weaker (Barone and Mocetti 2014). The problems often highlighted in this context, such as using emergency authority when it is not required/warranted or using it beyond the time it is required, abuse of discretionary authority, drawing up specifications to suit the firm desired to win the contract, restricting the number of bids, and caving in to political influences (Schultz and Søreide 2008: 523), have also been on display during the Covid-19 crisis.

There are of course compelling reasons to relax stringent procurement rules in emergencies to allow for a fast response proportional to the population´s needs. But such a lessening of oversight and ex ante checks must be compensated for by much more extensive ex post checks, that should be advertised widely to deter public officials from abusing discretion. Broadly, there are two main ways of strengthening ex post checks/monitoring.

Two Ways of Ex-post Monitoring

The first is to have complete and transparent documentation of all the contracts awarded and the related documents, a “keep the receipt” mentality and practice, and making these records publicly available as soon as possible. Several countries have been moving in this direction as a response to the crisis, often with the help of NGOs like the Open Contracting Partnership (The Economist 2020). Examples include Ukraine, that require the submission of a report for each contract within a day of its conclusion, which is then made publicly available on an internet platform; and as of 2016 a third of government contracts in Colombia were published on an e-procurement platform where they can then be scrutinized by the public. In the US, the user-friendly website USAspending.govprovide data on federal contracts, with advanced search functions including tags specific to Covid-19 contracting.

The organization Open Contracting Partnerships provide a list of suggestions for any government that is looking to increase transparency in procurement; it includes the timely publication of contracts, licenses, concessions, permits, grants, as well as related pre-studies and bid documents. A full list of best practices, which can be implemented at a low cost, can be found on their website (Open Contracting Partnerships 2020).

The second is to protect and incentivize whistleblowers. Adequate protection of whistleblowers is a first step, but protection is always partial and imperfect, and may therefore be insufficient to induce those close to frauds to come forward, given the terrible consequences they typically face (see e.g. Rothschild and Miethe 1999, Nyreröd and Spagnolo 2020c).

In the U.S., the False Claims Act (FCA), first enacted by President Lincoln to curb fraud on military supplies during the civil war, and strengthened in 1986, has gone one step further by providing whistleblowers with substantial monetary rewards when they report on procurement fraud. Building on the success of the FCA, the US has introduced similar programs in several areas, most prominently with respect to tax evasion (in 2006) and securities fraud (in 2011).

Providing meaningful monetary incentives to whistleblowers who report on particularly egregious frauds and corruption can have a substantial deterrent effect on potential fraudsters as several studies show (see e.g.  Wilde 2017, Johannesen and Stolper 2017, Wiedman and Zhu 2018, Amir et al. 2018, Leder-Lewis 2020; see Nyreröd and Spagnolo 2020a for a review of the earlier literature). Simple cost-benefit analysis shows that a well-designed and implemented whistleblower incentives scheme can be a highly cost-effective continuous monitoring tool for enforcement agencies and public prosecutors (see e.g. Nyreröd and Spagnolo 2020b).

As for the EU, it is conspicuously lagging behind. Even prior to the Covid-19 crisis there was a need for increased monitoring evidenced by a 2019 European Court of Auditors (ECA) report entitled “Fighting fraud in EU spending: action needed.” A central emphasis of this report is that the Commission lacks insight into the scale, nature, causes, and level of fraud, as well as the level of undetected fraud. In 2018 the EU adopted a Directive that would harmonize and strengthen whistleblower protection in the EU. While the new EU Directive on whistleblowing is a step in the right direction, it failed to provide a framework for whistleblower rewards.

This may have been a mistake, as standard detection methods, including whistleblower protections, have often proven inadequate. The recent Wirecard scandal is a testament to the failure of standard fraud detection methods. In June of 2020, the stock price of Wirecard dropped from €100 to sub €2 in less than nine days after it was revealed to be an Enron-level accounting fraud. The firm has also allegedly laundered money for mobsters and was involved in a range of shady practices. Since 2008, fraud accusations have been leveled several times against the firm and Wirecard´s response was to label their critics “market manipulators”. The German financial supervisors, instead of investigating Wirecard, went after those who correctly claimed that the firm was a fraud, including reporters at the Financial Times. This fraud went undetected for at least 12 years, costing investors millions and undermining trust in financial markets. Moreover, those correctly accusing Wirecard of fraud allege they were subject to harassment campaigns, including phishing attacks by hackers and intimidating surveillance outside their homes and offices [16]. This is perhaps not surprising given that Germany is a country with some of the worst protections for whistleblower [17].

The shortcomings of traditional methods of fraud detection may turn out to be especially costly and ineffective during the Covid-19 pandemic.

Conclusions

With increased public spending being a cornerstone of the response to this crisis, adequate monitoring of abuse of public funds will become more urgent. Some EU institution, such as the European Public Prosecutor’s Office, or the European Anti-Fraud Office, could be suitable for a whistleblower reward program, as investigators are likely stuck looking for needles in haystacks, or lack the necessary information to bring/recommend actions to recover funds. Irrespective of the lost opportunity of the Directive, evidence shows it is time to introduce serious (high stakes) whistleblower rewards programs in Europe, unless of course Europeans are not able to manage them, or are more interested in hiding rather than airing their dirty laundry.

References

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Whistleblowers During the Covid-19 Pandemic

Posted on June 15, 2020 | Policy Brief

Image with people standing in a building representing whistleblowers During the Covid-19 Pandemic

Numerous stories have emerged about whistleblowers being silenced and retaliated against during the Covid-19 pandemic. In this policy brief we consider some cases of retaliation against whistleblowers and cases illustrating the significance of the information they bring forward. Two facts about Covid-19 and whistleblowers become salient. First, it is hard to externally monitor behavior within care homes due to the risk of contagion (auditor-patient/patient-auditor). Second, it is hard to infer from outcomes (e.g. number of deaths) that management misbehaved, due to the high uncertainty and the many possible factors involved in the spread of Covid-19. Adequate whistleblower protections and confidential reporting channels are therefore essential to ensure transparency, compliance with safety rules, and more generally public accountability in the management of this crisis.

Whistleblowers are Silenced and Suffer Retaliation

The Covid-19 crisis has created pressure on governments, hospitals and secondary health institutions – in particular elderly care homes – to control the narrative on the spread of the virus and their response to it. As a result, we have already seen several whistleblowers being silenced and retaliated against.

The most (in)famous case is probably that of Li Wenliang, the Chinese doctor at Wuhan Central Hospital who warned his colleagues about a new SARS-like virus to on the 30th of December 2019. Four days later he was summoned to the Public Security Bureau where he was ordered to sign a letter in which he was accused of “making false comments” and “severely disturbed the social order”. Another seven persons were also arrested on suspicion of “spreading rumors”.

China is perhaps not the first country that comes to mind when considering adequate whistleblower protections, but the problem is a broad one.

In the US, several doctors and nurses have been fired and disciplined for expressing worries about their work conditions, also in relation to a lack of personal protective equipment (PPE). Nor is the issue of retaliation against whistleblowers localized to healthcare. The Vice President of Amazon’s cloud computing arm, Tim Brady, quit his job “in dismay at Amazon firing whistleblowers who were making noise about warehouse employees frightened of Covid-19”. Nine US senators also sent Amazon a request to explain its policy for firing workers after several employees who had expressed concerns over working conditions were laid off.

In Russia three doctors, two of which had protested their working conditions during the crisis suspiciously fell out of hospital windows, allegedly due to excessive work pressure. Two of the doctors died, and one doctor was threatened with criminal charges for spreading “fake news” about Covid-19. We do not know whether these cases were accidents, suicides, or retaliation for speaking up, but an investigation is currently ongoing.

The problem has been particularly severe in residential elderly care homes, where in many countries there have been extreme rates of contagion and deaths.

In Italy, complaints from caring personnel about lack of PPE and safety procedures in terms of restrictions in visits of relatives from ‘red zones’ and transfer of personnel and patients across departments, emerged already in the end of February. At the private elderly care home Trivulzio in Milan, caretakers claim that their early complaints were ignored by management, who allegedly also harassed those wearing face masks on the ground that they would scare guests. An investigation is currently ongoing, but if the allegations turn out to be true, numerous deaths in Lombardy could potentially be attributed to this negligence and could have been avoided.

In the UK, a country that had much more time than others to prepare for the arrival of the virus, a recent report by the whistleblower hotline Compassion in Care registers a dramatic increase in calls to their hotline: over 170 since the Covid-19 outbreak, while they normally receive no more than 30 cases per month. These whistleblowers at residential care homes, care agencies, and nursing homes continue to detail a widespread lack of protective equipment, and retaliation for raising these concerns. Five lost their jobs and are considering taking legal action.

In some countries there is instead a noticeable absence of whistleblowers at nursing homes and the like. Germany is one example. While this can be due to the country’s fast and apparently adequate response to Covid-19, the country also has an infamous history of mistreating whistleblowers, and the country´s protections are some of the weakest in the EU, which may have deterred potential whistleblower from reporting. A well-known example related exactly to nursing homes is the case of Heinisch vs. Germany, where a nurse was fired for reporting improper working conditions in 2005, and then lost her case for reinstatement at all levels of German labor courts, even though it was recognized that her claims were correct. She had to turn to the European Court of Human Rights to be vindicated, only after six years of legal hassle, though.

These are just some examples of the systemic issue of silencing and retaliation that is now emerging. Watchdog organizations are warning about a widespread and extensive mistreatment of whistleblowers worldwide during this pandemic. The Government Accountability Project details several cases of maltreatment of whistleblowers, describing the situation created by the Covid-19 crisis as “the largest attack on whistleblowers in the world”.

Other cases of whistleblowing, absent retaliation, further illustrate the crucial value of the information they bring forward. In Sweden for example, a country that should have been particularly careful given its softer approach to contain the virus, whistleblowers still reported a lack of PPE and poor safety routines in elderly care institutions. At one home, employees detailed how they went from caring for Covid-19 patients to caring for non-Covid patients while wearing inadequate safety protections. At that same home, it is estimated that more than 35 persons died from Covid-19: over a third of all residents.

Fighting Misinformation and Uncovering Wrongdoing

Protecting whistleblowers is also crucial to fight misinformation and fraud related to Covid-19. For example, a whistleblower recently alleged that the founder of JetBlue, who previously had argued against lockdowns, helped fund an influential yet controversial study which found that the infection rate in Santa Clara County, California, was 85 times higher than believed – which would have driven down the local fatality rate to flu levels at 0.12% – 0.2%. The whistleblower complaint also contained emails suggesting the authors of the study disregarded warnings raised by two other Stanford professors who attempted to verify the accuracy of the antibody tests used in the study.

Worries about abuse and fraud related to stimulus packages linked to Covid-19 have also been mounting. And indeed, the US Securities and Exchange Commission has seen a 35% increase in whistleblower claims received between mid-March and mid-May compared to the previous year.

There is already strong public support for whistleblower protections with respect to important matters like healthcare and elderly care (Butler et al., 2019), and as we have argued elsewhere (Nyreröd and Spagnolo, 2020a, 2020b), whistleblowers are currently not adequately protected or incentivized in the EU: they do not speak up to the degree desirable from a law enforcement/public interest point of view. The negative consequences of speaking up are often significant: blacklisting from the industry, harassment, and social and economic uncertainty are frequently associated with whistleblowing. This is not different with Covid-19 whistleblowers.

What Can Be Done

The state of whistleblower protection in Europe has been rather poor and uneven (Wolfe et al., 2014). In 2013, Transparency International rated a disappointing four countries in Europe as having “advanced” legal protection for whistleblowers. In recent years, several countries have enacted legislation to remedy the issue. France enacted Sapin II in 2017, which prohibits retaliation against whistleblowers; Sweden improved its protection in 2016 (Proposition 2015/16:128); and since November 2017, whistleblower protection in Italy, which was previously limited to the public sector, has been extended to the private sector.

It is only now, however, with the new EU Directive on Whistleblowing that we will see even protection levels for whistleblowers throughout the EU. Among other things the Directive would require firms with more than 50 employees to establish confidential internal whistleblower channels. The deadline for transposing the directive (implementing it into national law) is December 17, 2021.

EU member states should try to transpose the directive as soon as possible, as whistleblower protections are not only needed at nursing homes, but also at firms who may choose to put employees at excessive risk of infection when faced with high cost of compliance with safety measures. This is important, because monitoring compliance with safety measures externally will likely be difficult and costly, while the new directive contains several articles that would improve the informational flow within organizations but also externally to supervisory agencies.

To conclude, the Covid-19 crisis has created pressure to silence whistleblowers to control reputational risks for governments and private firms. If whistleblowers are successfully silenced, we risk ending up with an incomplete picture of the spread of the virus, a lack of public accountability, unnecessary deaths, and several good faith whistleblowers being retaliated against without adequate protection. Hastening the implementation of the new Whistleblower Directive is one way to ensure some level of protection for whistleblowers throughout the EU.

References

Nyreröd, T; and G Spagnolo, 2020a. “Myths and Numbers on Whistleblower Reward”, Regulation and Governance, forthcoming.

Nyreröd, T; and G Spagnolo, 2020b. “Financial Incentives for Whistleblowers: A Short Survey”, forthcoming in Cambridge Handbook of Compliance. Sokol, D., van Rooij, B. (Eds). Cambridge University Press.

Butler, J; D Serra; G Spagnolo, 2019. ”Motivating Whistleblowers”, Management Science, 66(2), 605-621.

Wolfe, S; M Worth; S Dreyfu; A Brown, 2014. “Whistleblower Protection Laws in G20 Countries, Priorities for Action.” Blueprint for Free Speech, The University of Melbourne, Griffith University, Transparency International Australia.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Covid-19: News for Europe’s Energy Security

Posted on May 7, 2020 | Policy Brief

Black and white image of natural gas processing plant representing Energy security Europe

While there has been a lot of attention on the effect of Covid-19-related developments in the oil market, the effect on the natural gas market has almost evaded media attention. For the EU, however, the gas market and especially the impact of the pandemic on the gas relationship with its largest gas supplier, Russia, is of high relevance. This brief discusses the potential implications of Covid-19 on this relationship both under the pandemic and during the expected slow economic recovery. We argue that, while in the short run the security of Russian gas supply is likely to improve, this is unlikely to be the case in the aftermath of the pandemic. To ensure gas supply security in post-pandemic markets, the EU may need to finally implement the long-awaited “speaking with one-voice” energy policy.

Introduction

The ongoing coronavirus pandemic will not only affect human lives, but also bring new economic and political challenges. The energy sector, and in particular the dramatic decrease of oil prices, has been in the news since the beginning of the Covid-19 crisis. But discussions have so far rarely touched the natural gas market, despite the pandemic taking its toll also on this market. As for oil, the demand and price have been negatively affected by the economic slowdown. While not as drastic as for oil, the price of natural gas in the EU has declined by approximately 40% since the beginning of 2020 (World Bank, 2020). However, the impact of the pandemic is likely to be quite different in oil and gas markets. There are multiple reasons for that; for example, oil and oil products are predominantly consumed by the transport sector while natural gas is mostly used in the power sector, the industry and households, and these sectors were differently affected by the Covid-19 pandemic.

Understanding the impact of the pandemic on the gas market is especially interesting from the European point of view, given that natural gas accounts for 25% of total energy consumption and two thirds of this gas is imported. The imports are also very concentrated, with the main supplier Russia providing around 40% of the gas, compared to 25% of the crude oil. This dependency, as well as a long history of tensions with third parties (Ukraine and Belarus) on the Russian gas transit routes, has made the EU’s concerns about the security of Russian gas supply much more pronounced than for oil (see Le Coq and Paltseva, 2012). The combination of these factors – i.e. the importance of natural gas for the EU and the long-standing concern about gas supply security warrant an analysis of the short and mid-term effect of the Covid-19 pandemic on the gas market, and, specifically, on the EU-Russia gas relationship. This brief discusses how the pandemic-driven decline in gas demand, and the potential shift in the balance of power between the parties may affect both the dependency on, and the transit of, Russian gas.

EU Dependency on Russian Gas Under the Covid-19 Pandemic

As is well known, Covid-19 and the associated lockdowns imposed by many EU Member States, have caused a slowdown in most economies and a decline in energy demand. However, for natural gas, the effect is likely to be significantly smaller than for oil. While we do not yet have statistics for the EU’s gas demand in recent months, the Norwegian energy consultancy Rystad Energy has predicted the decline of gas demand to be around 4% for March and April 2020. This forecast was given quite early in the course of the pandemic, and is very likely an underestimation; still, it is very different from the one for oil, with the demand drop estimated to be a whopping 34% in April.

One reason why we do not observe a sizable decrease in gas demand is that the natural gas is used in electricity generation, especially as a base-load fuel to compensate for the intermittency of green energy sources, such as sun and wind. With the reduced electricity demand, renewable power generation has become relatively more important in the electricity supply in many countries. Since mid-March 2020, the share of renewable power generation across the EU is 46%, nine percent higher than during the same period last year (Energy Transition Lab, 2020). Interestingly, in France, Germany, Belgium, the Netherlands, the Czech Republic, Poland and Hungary, the absolute volume of electricity generation by renewable sources even increased relative to the same period in 2019, despite declining energy demand. One potential channel, anecdotally recorded for Germany could be higher solar generation due to cleaner skies resulting from the decline in emissions because of lower fossil energy consumption. A higher volume of a renewable generation often requires more back-up power to maintain grid stability. While natural gas is not the only back-up source, this need might still limit the decline in gas demand (or even increase it like e.g. in the Czech Republic). Of course, cheaper gas prices may also play a role: for example, Slovakia and Romania experienced an increase in gas-based generation, but a drop in the renewable generation since mid-March 2020 relative to the same period in 2019. Finally, another reason for the moderate gas demand decline is its residential use – which is likely to be sustained due to the lockdown regime introduced by many countries.

When it comes to Russian gas imports, the official statistics since mid-March – roughly the beginning of lockdown policies across the EU – are not available yet. However, we can with some reservation look at the evolution of the volume of gas sales to the EU disclosed by Gazprom (2020). There was a very sizable decrease in Russian gas imports by the EU – of more than 21% – as compared to the same period last year but it started before the lockdown: January 2020 recorded a drop of 34% and February of 20%). This suggests that the current decrease in Russian gas imports is only marginally related to the pandemic, and more related to the overall gas market situation (such as relatively full gas storage in the EU in 2020, a warm winter, an increase in LNG imports, etc.).

It is, however, likely that the negative effect of the pandemic on Russian gas imports by the EU will be noticeably higher than it currently appears in the Gazprom data, thereby further decreasing the EU’s dependency on Russian gas. Moreover, since demand and prices decrease, substituting for Russian gas, were there a supply interruption, should be relatively easy and cheap with the current excess capacity of the natural gas market and the substantial storage in the EU.

Another reason for the improvement in the security of Russian gas supply to the EU is the observation that Russia’s dependency on oil and gas exports in combination with pandemic-associated factors may lead to a substantial economic downturn in Russia (Becker, 2020). In these dire circumstances, Russia is unlikely to further risk its gas export revenues by pursuing geopolitical goals through the means of gas supply and gas transit. For all these reasons, one may expect the security of Russian gas supply to the EU to improve during the pandemic.

However, the EU dependency on Russian gas may still be a concern due to medium-run effects of Covid-19. First of all, while the gas prices have been in decline for roughly a year now, the recent decrease in natural gas prices has accelerated the negative impact on the unconventional natural gas industry. For example, the US natural gas rig count has declined by 20% since mid-March 2020, which accounts for more than a third of the 54% year-to-year decline (Ycharts.com, 2020). Similarly, nearly 42% of Australian gas resources could be uneconomic under the current gas prices, according to Rystad Energy. While gas prices are unlikely to stay low forever, the industry will need time to recover even if/when the natural gas demand rises again. Moreover, the East-Asian markets are likely to be served first, as they are expected to recover from the pandemic shock before Europe. This dynamic, coupled with historically higher LNG prices in Asia may delay the LNG flows to Europe. A shortage of LNG in Europe, in turn, is likely to hinder any diversification strategy from Russian gas, weakening the EU’s bargaining power. The new Russia-China gas pipeline, “Power of Siberia”, operational since the end of 2019, will also be used to satisfy the post-Covid-19 Chinese gas demand which is likely to recover before demand picks up in the EU. Its use will then allow Russia to be less reliant on exporting gas to the EU, further contributing to the EU’s gas security concerns.

Transit of Russian Gas to the EU: Covid-19 Effect

The EU’s energy security also depends on the reliability of Russian gas transit to the EU. There are currently 5 transit routes connecting Russia to the EU (plus the routes that are serving the Baltic states and Finland without further transit), see Figure 1. Three onshore routes connect Russia to the EU via Ukraine and Belarus. There has been a history of gas transit disputes associated with these routes, at times threatening the Russian gas supply to the EU. Two newer offshore pipelines, Nord Stream 1 (in operation since 2011) and TurkStream (in operation since 2020) connect Russia directly to Germany, and to the South-East of Europe via Turkey. Further, one more offshore route to Germany, Nord Stream 2, is currently underway, with the operations announced to start in the first quarter of 2021. All three offshore projects are expected to not suffer from geopolitical transit issues.

In relation to the Covid-19 pandemic, there are likely to be two major effects on Russian gas transit. First, the inauguration of Nord Stream 2 is likely to be further delayed. Nord Stream 2 is 50% financed by Gazprom, and this financing scheme may be difficult to sustain after the fall in oil and gas prices and a significant decrease of Gazprom’s export revenues. Indeed, while the statistics for March and April 2020 are not yet available, the Russian customs statistics suggests that the USD value of gas exports from Russia in January-February 2020 has decreased by 45% relative to the same period last year. Because Nord Stream 2 could facilitate gas delivery to the EU in case of a transit conflicts, its expected delay may negatively impact the EU’s gas security.

Additionally, the Covid-19 related demand drop may impact the utilization of Russia-EU gas routes, driven by the current agreements between Russia and the transit countries. Russia and Ukraine have just signed a transit agreement for the next 5 years. This agreement was widely perceived as a diplomatic success of the EU (that facilitated the deal), given the historically difficult geopolitical relation between Ukraine and Russia. One of the new features of this agreement is of particular interest within the Covid-19 context. Unlike for previous deals, Russia agreed to prepay a fixed volume of gas transit, 178.1 mcm/day for 2020, and 110 mcm/day units for 2021-24 (Pirani et al., 2020). So, underutilization of this route is costly for Russia.

Figure 1. Gas supply Routes to the EU.

Source: Ukrainian Liaison Office in Brussels

With decreased demand due to Covid-19, warmer weather in the coming months and almost full gas storages in the EU, this contractual feature may affect how Russia allocates its gas exports across the routes. At least, in the short term, it may undermine Russian gas transit via the Belarus-Poland route. The concern about the utilization of this route in relation to the new Russia-Ukraine transit agreement has already been raised by Pirani et al. (2020). The Covid-19-associated decrease in gas demand is likely to make this concern much more real. Russia may use the Belarus-Poland pipeline sporadically, e.g. to adjust for the seasonal spikes in demand, without long-term capacity booking. Recent gas tensions between Russia and Poland (e.g. Poland winning in the arbitration court against Gazprom (RFE/RL, 2020), and Poland repeatedly expressing opinions and exercising legislative effort restricting the usage of Nord Stream 1 and construction of Nord Stream 2) may further exacerbate the issue.

In the medium term, however, when the EU gas demand has recovered but Nord Stream 2 is not yet in place, the Belarus-Poland route is likely to prove useful for Russia, at least starting from 2021 (when prepaid volumes of Russian gas transit via Ukraine will decline according to their agreement).

The transit contract between Russia and Poland is to be renewed in mid-May 2020, and as of now, it is unclear if, and how it will be written and whether the Belarus-Poland transit route will be used to a substantial degree or only marginally. If transit through the Belarus-Poland route is limited, it will imply poorer route diversification for a major part of European consumers of Russian gas, thereby lowering their security of Russian gas supply.   This may also put another strain on the bargaining power allocation within the EU and the EU’s intended common energy policy of “speaking with one voice” with external energy suppliers like Russia.

Conclusion

Summing up, the decrease in demand of natural gas, as well as other factors associated with the ongoing Covid-19 pandemic, such as economic recession and turbulence in stock markets, are likely to have noticeable implications for the security of Russian gas supplies to the EU in the short term. On the one hand, even if the current pandemic-associated decrease in demand of gas from Russia seems rather moderate, the ultimate negative effect on Russian gas imports by the EU is likely to be larger. Lower imports from Russia are likely to improve the security of supply, both through lower import dependency of the EU, and through improved market opportunities due to the current market’s overcapacity. On the other hand, in the medium run, lower demand also negatively affects the non-conventional gas industry, undermining the diversification opportunities to LNG, and, consequently, natural gas energy security. Further, a fall in the gas demand by the EU coupled with the newly signed transit agreement between Russia and Ukraine may potentially cause underusage of the Belarus-Poland transit route, thereby putting a strain on the diversification of Russian gas import routes to the EU and on the power balance within the EU.

Energy security might be even more of a concern in the post coronavirus period when the economy is slowly recovering, and cheap and guaranteed energy supply is crucial. To ensure this supply, national efforts combined with an EU-wide policy coordination would be required. The long-discussed “speaking with one voice” common energy policy may finally need to materialize in order to facilitate reliable access to natural gas.

References

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Gender in Economics: From Survival to Career Opportunities

Posted on January 14, 2019 | FROGEE PROJECT, Policy Brief

20190121 Gender in Economics Image 01

Gender inequality goes beyond discrimination and sexism. It is also a matter of efficiency and development, and therefore, the socioeconomic losses that result from such inequality must be acknowledged and tackled. This policy brief summarizes the presentations held during the 6th SITE Academic Conference at the Stockholm School of Economics on December 17-18 2018. The event brought together scholars from around the world to examine existing forms of gender inequality, its causes, consequences, and policy interventions through a series of keynote speeches, research presentations and panel discussions.

Gender and survival

The reality of gender inequality is diverse throughout the world. The extent to which women and men face different opportunities and reach different outcomes vary substantially across countries and regions, and the forms of inequality that women face also vary geographically.

While richer countries have mostly closed their gender gaps in health and education, in other parts of the globe women are still struggling to survive, to make their marriage and reproductive choices freely, and to achieve the same educational opportunities as men. This is exactly where modern economic research can facilitate the understanding of the roots of such inequalities in each society, as well as the most likely drivers of change.

Corno, Hildebrandt and Voena (2017) show that in Sub-Saharan Africa and India, the age of marriage is a result of short-term changes in economic conditions (such as a reduction in crop yields due to droughts). Therefore, through for instance insurance mechanisms and temporary transfers, economic policy can influence marriage markets and the age of marriage. Relatedly, according to Ashraf, Bau, Nunn and Voena (2018), a girl in Indonesia or Zambia has a higher probability of being educated if she belongs to a group practicing bride price, defined as the “price” paid by a groom or his family to the bride’s family. This means that marriage markets could be a driver of educational investment. Cousin marriage is another issue within this context. Edlund (2018) suggests that this system serves as a barrier for economic growth by favoring men over women, the old over the young, and the collective over the individual. In general, challenging these marriage systems and improving female economic opportunities require a deeper understanding of the economic role of traditional cultural norms and institutions.

Some groups of women struggle for survival even in the so called “developed world”, being victims of gender violence. Sex workers in the United States are a particularly vulnerable population in this matter. Cunningham, DeAngelo and Tripp (2017) point out that, given that prostitution in most cities of the US isn’t only illegal, but also very dangerous (recording the highest homicide rate of any female occupation), it is critical to improve sex workers’ safety. Craigslist Erotic Services (CES) seemed to have contributed to it, by reducing female homicide rates by 17.4%. Apparently, this was a result of street prostitutes moving indoors and being able to filter clients to be safer. It is, therefore, suggested that the closure of such a platform put sex workers in an even more vulnerable position. Similarly, when it comes to adult entertainment establishments and its relation to sex crimes, Ciacci and Sviatschi (2018) argue that this type of businesses helps decrease daily sex crimes between 7-13% in the precinct where they are located.

When discussing approaches to prostitution, the “Nordic Model” has been highly praised and adopted by several countries. The term refers to a reform initiated in Sweden that considers buying sex a criminal offense, while decriminalizing those who are prostituted. However, preliminary results from Perrotta Berlin, Spagnolo, Immordino and Russo (2018) suggest that intimate partner violence and violence against women might have increased because of its enactment in Sweden.

Gender violence, however, isn’t only domestic or affecting sex workers. Borker (2018) claims that, in India, female college students are willing to choose less prestigious universities, to make additional expenses and to spend more time on transportation than their male counterparts only to avoid harassment on the street or public transportation. Street harassment, therefore, perpetuates gender inequality in both education and potentially the labor market.

Challenging social norms

As already seen, even the most gender-equal countries still suffer from persistent forms of inequality that need to be acknowledged and tackled. Doing so will result both in fairer societies and in more efficient economies, because it will make full use of both halves of the world’s skills and knowledge.

Friebel, Auriol and Wilhelm (2018) state that, in Europe, it is harder for women to make a career in economics. The representation of women in academics is low, and the higher ranked the university, the lower is the representation. This could be a consequence of several issues, one of them being the “glass ceiling”.

The glass ceiling, according to Bertrand (2017), is the phenomenon by which women remain underrepresented in high-level occupations, and earn less. Even in countries such as Denmark and Sweden, women still receive less for the same jobs. There are many potential explanations for this. One of them refers to the gender differences in psychological attributes in work, such as the idea of women performing worse under pressure or being unwilling to compete. This interpretation ultimately falls under the nature vs nurture discussion and only accounts for up to 10% of the pay gap. Another reason states that women suffer the penalties associated with demanding more flexibility. Such demand comes from the need to perform non-market work, like domestic work and, especially, caring for children. This means that women, especially the more educated ones, are paying a disproportionate price in the labor market for raising a couple’s children. Giving women more flexibility won’t crack the glass ceiling, au contraire, it will backfire because flexibility is negatively priced in the market. Besides, it doesn’t address the earning gaps. A more compelling proposal is to shift the focus from increasing flexibility to changing social norms and gender role attitudes. Normalizing and encouraging paternal child care in workplaces, for example, could be a way to do so.

Social norms based on traditional gender stereotypes also seem to be the reason why in Sweden, promotions to top jobs dramatically increase women’s probability of divorce but do not affect men’s marriages, as reported by Folke and Rickne (2018). In this case, promoting norms and policies with a more gender-equal approach to couple formation could increase the share of women in top jobs.

Given the importance of social norms, understanding how they can change is crucial. In Saudi Arabia, two studies were conducted on the influence of misperceived social norms. Both showed that the low-cost intervention of simply providing information could make a big difference. In one case, Bursztyn, González and Yanagizawa-Drott (2018) have evidenced that most young married men privately support female labor force participation (FLFP) outside of home. Nevertheless, they tend to underestimate the level of support for FLFP by other men. When correcting those misperceptions, the men’s willingness to let their wives join the labor force increases. Comparably, Ganguli and Zafar (2018) have shown that there is an increased likelihood of working full-time for female students when they, along with their close circles, receive information about the labor market and the aspirations of other women peers.

Challenging social norms isn’t only beneficial when discussing the glass ceiling and FLFP, it also has the potential to improve public health. In fact, Milazzo (2018) argues that women’s increased mortality rate in India can be an unintended consequence of son preference. Son preference induces women with a first-born daughter to adopt behaviors that increase the risk of maternal morbidity and mortality. Therefore, interventions to change deeply rooted social norms such as the boy preference could significantly reduce maternal mortality risk.

Bridging research and policy

In Malawi, research by Perrotta Berlin, Bonnier and Olofsgård (2017) on aid project location suggests that proximity to aid has a positive effect on the lives of women and children. Likewise, Goldstein (2018) reports that the World Bank’s Empowerment and Livelihoods for Adolescents (ELA) program in Uganda has also led to positive reproductive outcomes and income effects. These results illustrate the importance of reducing the divide between research and policy. Research has the potential of serving as an instrument for informed policy-making and aid intervention.

The Organization for Economic Cooperation and Development (OECD), for instance, applies research to create tools that help improve economic and social well-being. Two of those tools are the Social Institutions and Gender Index (SIGI) and the Development Assistance Committee (DAC) Gender Equality Policy Markers. On one hand, Missika (2018) explains that the SIGI is a cross-country measure of discriminatory social institutions against women and girls. Though the progress is slow (it might take around 200 years to close the gender gaps), its use gradually promotes the creation of locally designed solutions that, combined with adequate legislation, could enhance gender equality. On the other hand, Williams (2018) states that the DAC Gender Equality Policy Markers are meant to ensure that women have access to and benefit from finance.

Consistently , for the Swedish International Development Agency (SIDA), which works on behalf of the Swedish government, gender equality is a priority that permeates its interventions. In this context, the Feminist Foreign Policy has strengthened Sweden’s commitment in the topic.

Prior to finalizing the conference, representatives of the FROGEE Network (Forum for Research on Gender Economics in Eastern Europe and Emerging Economies) made a short presentation about the key challenges for achieving gender equality in their countries and the research opportunities available.

Conference material, including presentations, can be found here.

Speakers at the conference

Marianne Bertrand, University of Chicago

Alessandra Voena, University of Chicago

Alessandra González, University of Chicago

Anders Olofsgård, SITE

Annamaria Milazzo, World Bank

Bathylle Missika, OECD Development Centre

Eva Johansson, SIDA

Girija Borker, World Bank

Guido Friebel, Goethe University Frankfurt

Ina Ganguli, University of Massachusetts

Amherst Johanna Rickne, Stockholm University

Lena Edlund, Columbia University

Lisa Williams-Katz, OECD

Maria Perrotta Berlin, SITE

Markus Goldstein, World Bank

Michal Myck, CenEA

Riccardo Ciacci, The University Loyola Andalucía

Scott Cunningham, Baylor University

References

  • Ashraf, Nava; Natalie Bau, Nathan Nunn, and Alessandra Voena. 2018. “Bride Price and Female Education”. The National Bureau of Economic Research Working Paper No. 22417.
  • Bertrand, Marianne. 2017. “The Glass Ceiling”. Becker Friedman Institute for Research in Economics Working Paper No. 2018-38.
  • Borker, Girija. 2018. “Safety First: Perceived Risk of Street Harassment and Educational Choices of Women”. Job market paper.
  • Bursztyn, Leonardo; Alessandra González, and David Yanagizawa-Drott. 2018. “Misperceived Social Norms: Female Labor Force Participation in Saudi Arabia”.
  • Ciacci, Riccardo; and Maria Micaela Sviatschi. 2018. “The Effect of Adult Entertainment Establishments on Sex Crime: Evidence from New York City”.
  • Corno, Lucia; Nicole Hildebrandt, and Alessandra Voena. 2017. “Age of Marriage, Weather Shocks, and the Direction of Marriage Payments”. The National Bureau of Economic Research Working Paper No. 23604.
  • Cunningham, Scott; Gregory DeAngelo, and John Tripp. 2017. “Craigslist’s Effect on Violence Against Women”.
  • Edlund, Lena. 2018. “Cousin Marriage Is Not Choice: Muslim Marriage and Underdevelopment”. American Economic Association Papers and Proceedings, Volume 108, pages 353- 57.
  • Folke, Olle; and Johanna Rickne. 2018. “All the Single Ladies: Job Promotions and the Durability of Marriage”.
  • Friebel, Guido; Emmanuelle Auriol, and Sascha Wilhelm. 2018. “Women in Europen Economics”. [Mimeo]
  • Ganguli, Ina; and Basit Zafar. 2018. “Information and Social Norms: Experimental Evidence on Labor Market Aspirations of Saudi Women”. [Mimeo]
  • Goldstein, Markus. 2018. “Evidence on adolescent empowerment programs from four countries”. [Mimeo]
  • Milazzo, Annamaria. 2018. “Why are adult women missing? Son preference and maternal survival in India”. Journal of Development Economics, Volume 134, pages 467-484.
  • Missika, Bathylle. 2018. “Are laws and social norms still an obstacle to gender equality? Lessons from the SIGI 2019”. [Mimeo]
  • Perrotta Berlin, Maria; Evelina Bonnier, and Anders Olofsgård. 2017. “The donor footprint and gender gaps”. WIDER Working Paper 2017/130, United Nations University World Institute for Development Economics Research.
  • Perrotta Berlin, Maria; Giancarlo Spagnolo, Giovanni Immordino, and Francesco Flaviano Russo. 2018. “Prostitution and Violence: Empirical Evidence from Sweden”. [Mimeo]
  • Williams, Lisa E. 2018. “Financing for gender equality beyong ODA”. [Mimeo]

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

How Should Policymakers Use Gender Equality Indexes?

Posted on November 13, 2018 | Policy Brief, Policy Papers

We look at the development of gender inequality in transition countries through the lens of the Gender Inequality Index (GII), which aims to capture overall gender inequality. Extending the measure back to 1990, we look at the development of the overall index as well as that of its components. We show that, even though gender inequality in transition countries for the most part has decreased since 1990, once overall development is taken in account these countries appear to fare better in 1990 than today. We also caution against relying exclusively on composite indexes to understand patterns of gender inequality. While the desire of policy makers to get one number that captures gender inequality development is understandable, weak correlations of the GII with other indexes (over years when multiple gender inequality indexes exist) as well as across sub-indexes suggests that such an approach has limitations. Finally, we emphasize the need to understand levels as well as trends and underlying mechanisms to better inform policy to improve gender equality.

On Measuring Progress

When studying economic development, or any issue really, one faces the challenge not only of finding the right way to identify and measure what are often complex changes, but also of communicating the bottom line efficiently. This naturally leads to the search for a single metric according to which we can rank progress and follow it over time. In the realm of economic development the standard measure is GDP growth. But, of course, focusing only on GDP leaves out many important dimensions of development, such as health and education.[1] In an attempt to capture these dimensions, while still arriving at a single number that measures development, the Human Development Index (HDI) was developed in the late 1980s. Since then, a number of alternative indexes capturing additional aspects of human wellbeing have been suggested; see the report by the “Commission on the Measurement of Economic Performance and Social Progress” (Stiglitz, Sen and Fitoussi, 2009).

Just as for overall development, there is great interest in single measures that capture the gender dimension of this development. Over the past decades a number of such “gender equality indexes” have been developed by international organizations such as the UNDP, the EIGE (European Institute for Gender Equality) and the WEF (World Economic Forum), to name a few.

These measures receive a lot of attention and in particular the reporting of country rankings tends to have an influence on political and policy discussions. The various indexes proposed differ in what dimensions they include (as will be explained below) and, much as a consequence of this, in the time periods they can cover. In some cases (as will also be shown below) it is possible to extend the time coverage of the indexes, but most of the times it is hard to recover the underlying data.

In this brief we summarize what the most popular indexes tell us about the development of gender equality in transition countries, contrasting these to Western European countries.[2] Whenever we have been able to find the underlying data, we also add to publicly available measures by extending indexes back to early 1990s. We then comment on the development of gender equality in transition countries and, perhaps most importantly, on why an indexes-based analysis should be interpreted with some care.

Gender Equality Before 1990

As has often been pointed out, the Soviet Union and many of the countries in Eastern and Central Europe were, at least in some dimensions, forerunners in terms of promoting gender equality (e.g., Brainerd, 2000; Pollert, 2003; Campa and Serafinelli, 2018). This was mainly due to the high participation of women in the labor market as well as the (official) universal access to basic health care and education.

However, some scholars have suggested that not all aspects of gender equality were as advanced in the countries in the Soviet Union and in Central and Eastern Europe (Einhorn, 1993; Wolchik and Meyer, 1985). Even though women were highly integrated in the labor market, they were also still expected to take care of child rearing and house work (UNICEF, 1999). The gender pay gap and gender segregation in the labor market was also similar to levels found in OECD countries. In addition, despite the high number of women in representative positions in communist party politics, women were rarely found in positions of real power in the political sphere (Pollert, 2003).

Generally speaking, while the communist regimes succeeded in promoting women’s access to the labor market and tertiary education, they failed to eliminate patriarchy (LaFont, 2001). Such a dichotomy gives rise to a broad set of questions regarding gender equality in transition countries as well as the measurement of gender equality in this context. What happened to gender equality, in relation to economic growth, during the transition, when new governments often broke with the tradition of promoting women’s employment and education? Did gender equality enhanced by communism leave a legacy or did underlying patriarchic values characterizing many of the communist societies come to dominate? How should we regard developments of indexes that try to weight several components within a context, such as that of transition countries, where these components may move in different directions from each other, given the dichotomy characterizing gender relations?

The Different Indexes

There are several different indexes that are often quoted in policy discussions. Two important measures are the Gender Development Index (GDI) and the Gender Inequality Index (GII), both calculated by the UNDP and reported annually in the Human Development Report (HDR). A third, more recent index that has received increasing attention is the World Economic Forum’s global Gender Gap Index (GGI), which is published in the yearly Gender Gap Report. These three can serve as illustrations of what gender equality indexes typically try to capture.

The Gender Development Index (GDI) essentially measures gender differences in the Human Development Index (HDI). The HDI in turn aims to capture achievements in three basic dimensions of human development: health (measured by life expectancy), knowledge (measured by expected and mean years of schooling) and living standards (measured by GNI per capita). The GDI then basically tries to assess the relative performance in these three dimensions for men and women respectively. If health (or education, or income)  in the population on average goes up, this improves the HDI. But to the extent that the improvements are felt differently by men and women, this will show in the GDI. There are several potential problems with the measurement of this index, especially when it comes to dividing GNI per capita between men and women (see e.g. Dijkstra and Hanmer, 2000); on the other hand, the index offers a transparent way to connect gender inequality to the HDI measure.

The other UNDP measure, the Gender Inequality Index (GII), was reported for the first time in the 2010 Human Development Report. It was created to address some of the perceived shortcomings of its forerunner, the Gender Empowerment Index (GEM) which had been introduced together with the GDI in 1995 (see e.g., Klasen and Schuler, 2011 for problems with GDI as well as GEM). The GII measures gender inequalities in three dimensions of human development: 1) reproductive health, measured by maternal mortality and adolescent birth rates; 2) empowerment, measured by representation in parliament and secondary education among adults; and 3) economic status, measured by labor force participation. As with the GDI, the areas of health, education, and economic empowerment are present, but the index also considers some aspects of health that are more directly relevant for women, and includes a component trying to capture political participation. The economic measure of labor force participation is also somewhat easier to interpret (and measure) than GNI divided between men and women. As for the GDI, GII country-values from 1995 are available on the UNDP website.  Conveniently for our purpose, most of the underlying data that the index is based on are also made available from the UNDP for the years 1990, 1995, 2000, 2005, and every year between 2010 and 2015, with the only exception of female seat share in parliaments in 1990[3]. We downloaded the latter from the World Bank indicators database[4]. We also added information on the share of women in the 1990 Polish Parliament, from the Inter-Parliamentary Union[5], and on the share of women in the 1990 Georgian “Supreme Council,” from Beacháin Stefańczak and Connolly (2015).

A third more recently developed index is the Global Gender Gap Index. This covers areas of political empowerment, health and survival, economic participation and educational attainment, as measured using 14 different variables. An indicator is available for each of the sub areas covered, which are then weighted together in an overall indicator of the gender gap. The Global Gender Gap Index is clearly more detailed and provides a more nuanced picture of existing gender gaps compared to the GDI or the GII. But this amount of detail also comes with potential costs; it is more difficult to interpret the overall index as there are more underlying components that may change simultaneously, and it is also more difficult to reconstruct the index back in time.

What Does the GII Index Tell Us About Gender Equality in Transition Economies?

Among the above mentioned indexes, we focus on the GII here. Extending this measure when possible allows us to study gender inequality starting from 1990 for a limited set of countries (we expand the sample of countries when looking at different dimensions of the GII separately below)[6]. Figure 1 reports values for the index in box plots, which show the index median, maximum, minimum, 75th and 25th percentile for two groups of countries: transition countries and Western-European countries. When interpreting Figure 1, recall that higher GII values imply more inequality.

Figure 1. The Gender Inequality Index in transition countries and Western Europe, 1990-2015

Source: Own calculations based mainly on UNDP data.

Figure 1 shows that based on the GII, median gender inequality is larger in transition countries than in Western Europe and has been so throughout the entire period since 1990. In both regions the index shows a decreasing trend, after an initial increase in 1995 in the transition countries. Below we will show that this is mainly due to a drop in female representation in national parliaments. The variance of the index scores has declined over time in Western Europe, while it remained mostly unchanged in the transition countries[7].

This first piece of evidence from the data is somewhat at odds with the common notion that transition countries enjoy relatively low level of gender inequality. However, two qualifications are in order here. First, transition and Western European countries are generally at different levels of development. Figure 2 displays the country groups performance in relation to their level of human development. This is done by measuring the difference between their GII ranking and their HDI ranking among all the countries with non-missing GII values in the years considered. The larger the difference, the worse the group performance in terms of gender inequality in relation to its level of development.

Figure 2. Difference between Gender Inequality Index ranking and Human Development Index ranking in transition countries and Western Europe, 1990-2015

Source: Own calculations based mainly on UNDP data.

The trends of transition countries and Western Europe are now opposite. In the former group, in 1990 the median standing in terms of gender equality was better than that in human development; this difference appears to have narrowed over time, and it is close to zero in 2015. Western European countries have instead improved their gender equality in relation to their level of overall human development over the period studied. Put differently, the gains in human development made by former socialist countries since the transition have not translated into comparable gains in gender equality as measured by the GII index.

Second, it is also important to emphasize that, as noted above, according to several scholars the socialist push in favor of gender equality was directed only to certain spheres of women’s lives, namely their economic empowerment. This suggests that a composite index can mask important contrasting patterns among its components.

In Figures 3 to 5 we document that different variables indeed paint quite diverging pictures of gender inequality in transition countries.

Figure 3. Development of adolescent births and maternal mortality, 1990-2015

Figure 4. Development of secondary education and share of women in parliament, 1990-2015.

Figure 5. Labor force participation, 1990-2015

Source: Own calculations based mainly on UNDP data.

In each figure we display box-plots for the three areas covered by the GII: health (measured by teenage births and maternal mortality), empowerment (measured by secondary education and share of women in Parliament) and labor force participation. Looking at the different variables separately also allows us to increase the number of countries significantly, since for many countries only the seat share of women in parliaments is missing in 1990.

As the figures show transition countries in 1990 displayed relatively low levels of gender inequality in labor force participation and secondary education. Over the last 25 years, they have kept improving the latter, while the former has stalled, resulting in Western European countries displaying a higher median level of gender equality in labor force participation for the first time around 2010. Reproductive health, while improving since the transition, is still far from converging to Western European standards. Finally, political representation appears to be responsible for the increase in inequality immediately after the transition that we have noted in Figure 1. While it is hard to compare the meaning of representation in the context of 1990 totalitarianisms to that of the democratic regimes emerged later, during the regime change women de facto lost descriptive representation, which was sometime guaranteed in socialist times by gender quotas (Ostrovska, 1994).

In conclusion, breaking down the GII by its components shows that, while Western European countries have invariantly improved their levels of gender equality since 1990, the trend in transition countries depends on the measure one looks at: women maintained but did not improve their relative status in the labor force, they gained more equality in education and especially in terms of reproductive health, and lost descriptive political representation.

What Does the GII Index (And Other Indexes) Not Tell Us?

The conclusion in the previous paragraph raises the question of which other areas of progress, stagnation or deterioration in gender equality in transition countries that might be overlooked in the GII index. Above, we have summarized two more indexes, the GDI and the Gender Gap Index, which focus on additional dimensions of gender inequality but are more limited in terms of time availability. For the time over which there is overlap between the available indexes, the correlation between the GII index and the GDI and the Gender Gap Index respectively, is roughly 0.60. Interestingly, such correlation is higher in the sample of western European countries (0.64 and 0.68 respectively); when the sample is limited to transition countries, the correlations are down to 0.40 and 0.50 respectively.

Several factors might account for the differences across indexes. Unlike the GII, both the GDI and the Gender Gap Index, for instance, include measures of income inequality. On the other hand, the GDI, as pointed out above, does not account for issues related to reproductive health and political representation. The Gender Gap Index is the only one to include, among the health measures, sex-ratios (typically defined as the ratio of male live births for every 100 female births). This turns out to be especially important for some of the transition countries: in the most recent Gender Gap Report, Georgia, Armenia and Azerbaijan remain among the worst-performing countries globally on the Health and Survival sub-index, due to some of the highest male-to-female sex ratios at birth in the world, just below China’s. This goes hand in hand with very high scores in terms of gender equality in enrolment in tertiary education, for which each of these countries ranks first (at par with a few other countries), having completely closed the gender gap. In fact, women are more likely to be enrolled in tertiary education than men.

The relatively low correlation among the different indexes for the group of transition countries also deserves special attention, because it might be a direct consequence of the peculiar history of women’s rights and empowerment in the region. Since some dimensions of gender equality were fostered through a top-down approach, rather than as the result of demands and needs expressed by an organized society, it is more likely that over the last thirty years elements of modernization coexisted with more traditional forms of gender inequality.

Finally, it is worth pointing out that none of the above indexes accounts for important dimensions of gender inequality such as,: gender violence, division of chores in the household, political representation at the local level, and the presence of women in STEM’s professions (where the largest job creation might happen over the next couple of decades). Once more, some of these measures might be particularly relevant for transition countries. Just to mention one example, gender violence is an urgent issue in a few of the countries in the area[8]. A case in point in this respect is Moldova: in 2017, the country ranked 30th out of 144 countries in the Gender Gap Index. Its rank for the sub-index called “Economic Opportunity and Participation” was 11[9]. The country performs especially well in terms of economic opportunity and participation because women not only participate in the labor market in almost equal rates as men, but they are also relatively fairly represented in professions traditionally less feminized elsewhere, such as “professional and technical workers” and “legislators, senior officials and managers.” At the same time, gender violence appears quite prevailing in Moldova: according to the UN, in 2014 “lifetime prevalence of psychological violence” in Moldova was of 60%. Official country statistics also report that the percentage of ever-partnered women aged 15-65 years experiencing intimate partner physical or sexual violence at least once in their lifetime in 2011 was 46%[10].

While limited in scope, the example above illustrates how some of the available indexes might not capture some important drivers of gender inequality in the region.

Conclusion

In this policy brief, we have reviewed some of the available gender inequality indexes that are commonly used in policy discussion as well as in policy-making.

We have then discussed gender inequality in transition countries focusing on one of these indexes, the Gender Inequality Index, whose span we have extended to the beginning of the transition period. Our analysis has highlighted some points to be mindful of when using comprehensive indexes to discuss gender inequality, especially in transition countries:

  • It can be fruitful to analyze gender inequality indexes in relation to levels of development. Some issues related to gender inequality, such as maternal mortality, are potentially addressed with a comprehensive strategy aimed at overall development. Conversely, other drivers of gender inequality, such as women’s political empowerment or gender violence, might require more targeted policy interventions, since they do not necessary go hand in hand with overall development.
  • While comprehensive indexes can be useful in terms of effective communication, it is often difficult to compress all the potential forms that gender inequality can take into a single index, especially over time. This is due to both conceptual issues and data limitations. Moreover, even when this is done, a comprehensive index can overshadow important sources of gender inequality if it is composed of sub-indexes that move in opposite directions.
  • The previous point can be especially relevant in the context of transition countries, which historically experienced a top-down approach to gender equality, the results of which in the long-term appear to be major advancements in some dimensions of women’s empowerment and contemporary potential backlash in other dimensions. In the context of transition countries, for instance, it has been argued that low levels of female representation in political institutions can be the result of women’s large participation to the labor market while division of roles in the household remained traditional. In the words of anthropologist Suzanne LaFont, “Women have been and continue to be overworked, and their lives have been over-politicized, the combination of which has led to apathy and/or the unwillingness to enter the male dominated sphere of politics. Many post-communist women view participation in politics as just one more burden.”[11] In such a context, average values of an index on gender equality might mask high achievements in economic empowerment coexisting with lack of political representation.
  • Identifying policies to address gender inequality in transition countries might be especially difficult because, depending on the dimension that one focuses on, the challenge at hand is different: in terms of education and employment, the policy goal appears to be maintaining current levels of equality or increasing them from relatively high initial points; the type of policies to do so are likely different than those used in Western European countries in the last 30 years, where the challenge was rather how to increase equality from relatively much lower levels. Conversely, in other dimensions the challenge is how to make major leaps forward, which move transition countries closer to Western European standards: this is the case for sex-ratios, for instance, and reproductive health more in general. The importance of initial levels and trends for policy implications also showcases how crucial it is to acquire more historical knowledge of policies, institutions, and statistics.

Overall, policy discussions and policy-making should go beyond mere descriptions of what indexes and related international comparisons tell us about gender inequality. A better knowledge and understanding of all of the drivers of gender inequality, of their historical evolution, and of their connections both with overall development and among them, is crucial to give sound policy recommendations.

References

  • Beacháin Stefańczak, K.Ó. and Connolly, E.(2015),  ‘Gender and political representation in the de facto states of the Caucasus: women and parliamentary elections in Abkhazia’. Caucasus Survey, 3(3), pp.258-268.
  • Brainerd, E. (2000), ‘Women in Transition: Changes in Gender Wage Differentials in Eastern Europe and the Former Soviet Union’, Industrial and Labor Relations Review, 54 (1), pp. 138-162.
  • Campa, P. and Serafinelli, M. (2018), ’Politico-economic Regimes and Attitudes: Female Workers under State-socialism’, Review of Economics and Statistics, Forthcoming.
  • Dijkstra, A. and L. Hanmer (2000), ‘Measuring socio-economic gender inequality: towards an alternative for UNDP’s Gender-related Development Index’, Feminist Economics, Vol. 6, No. 2, pp. 41-75.
  • Einhorn, B. (1993), Cinderella goes to market: citizenship, gender, and women’s movements in East Central Europe, London: Verso.
  • Klasen, S. and Schuler, D. (2011) Reforming the Gender-Related Development Index and the Gender Empowerment Measure: Implementing Some Specific Proposals. Feminist Economics. (1) 1 – 30
  • LaFont, Suzanne (2001), ‘One step forward, two steps back: women in the post-communist states.’ Communist and post-communist studies 34(2), pp. 203-220.
  • Ostrovska, I. (1994). Women and politics in Latvia. Women’s Studies International Forum 2, 301–303.
  • Pollert, A. (2003), ‘Women, work and equal opportunities in post-Communist transition’, Work, Employment and Society, Volume 17(2), pp. 331-357.
  • Stiglitz, Joseph, Amartya Sen, and Jean-Paul Fitoussi (2009). `The measurement of economic performance and social progress revisited.’ Reflections and overview. Commission on the Measurement of Economic Performance and Social Progress, Paris.
  • Tur-Prats, Anna (2018). Unemployment and Intimate-Partner Violence:  Gender-Identity Approach. GSE Working Paper No. 1564
  • Unicef. Women in transition. 1999.
  • UN. The World’s Women 2015.
  • Wolchik, S. L. and Meyer, A.G. (1985), Women, State and Party in Eastern Europe, Durham, NC: Duke University Press.

Footnotes

  • [1] In contrast to a common perception, economists are generally well-aware of the limitations of GDP as a measure of welfare. In fact, the reference manual of national accounts, the SNA 2008, makes this explicit in stating that there is “no claim that GDP should be taken as a measure of welfare and indeed there are several conventions in the SNA that argue against the welfare interpretation of the accounts”.
  • [2] By “transition countries,” we refer to all countries that were part of the Soviet Union plus the Central and Eastern European countries that were heavily influenced by the Soviet Union before 1990 (not including Albania and former Yugoslavia). Starting from this, we – as will be made clear below – sometimes limit the set of countries further depending on data availability.
  • [3] http://hdr.undp.org/en/data
  • [4] https://data.worldbank.org/indicator/SG.GEN.PARL.ZS
  • [5] http://archive.ipu.org/parline-e/reports/2255_arc.ht
  • [6] For Western Europe these countries are: Austria, Belgium, Cyprus, Denmark, Finland, France, Greece, Iceland, Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Sweden, and Switzerland. The transition countries are: Armenia, Bulgaria, Georgia, Hungary, Poland, Romania, Russian Federation.
  • [7] The outlier among Western countries is Malta.
  • [8] While explaining the sources of gender violence in the region is beyond the scope of this report, incidentally we notice that, according to recent research, female economic empowerment in a context where patriarchal values are dominant might backfire against women in the form of increased gender violence. See Tur-Prats, 2018.
  • [9] http://reports.weforum.org/global-gender-gap-report-2017/dataexplorer/#economy=MDA
  • [10] UNFPA (2015). Combatting Violence against Women and Girls in Eastern Europe and Central Asia. https://eeca.unfpa.org/en/publications/combatting-violence-against-women-and-girls-eastern-europe-and-central-asia
  • [11] LaFont, Suzanne (2001). One Step Forward, Two Steps Back: Women in the Post-Communist States. Communist and Post-Communist Studies, Vol. 34, pp 208.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Gender Gaps in Transition – What do we learn (and what do we not learn) from gender inequality indexes?

Posted on November 12, 2018 | Policy Brief

20181112 Gender Gaps in Transition Image 01

We look at the development of gender inequality in transition countries through the lens of the Gender Inequality Index (GII), which aims to capture overall gender inequality. By extending the measure back to 1990, we show that even though gender inequality in transition countries for the most part has decreased since the fall of the iron curtain, once overall development is taken into account, transition countries did better in relation to other countries in terms of rank differences before transition. We, however, caution against relying exclusively on composite indexes to understand patterns of gender inequality. While the desire of policy makers to get one number that captures gender inequality development is understandable, weak correlations across different overall indexes, as well as across different sub-indexes that make up each index, suggest that such an approach has limitations.

Indexes of gender inequality

In the public debate of socio-economic issues there is an understandable interest in single measures that summarize complex issues, describe historical developments and allow international comparisons. The use of GDP to measure economic development is the most immediate example of this way of proceeding. The same applies to gender inequality. Over the past decades a number of “gender equality indexes” have been developed by international organizations such as the UNDP, the EIGE (European Institute for Gender Equality) and the WEF (World Economic Forum), to name a few. These measures receive a lot of attention and in particular the reporting of country rankings tends to have an influence on political and policy discussions.

In this brief, we study the development of the Gender Inequality Index (GII) in transition countries, contrasting these to Western European countries.  By transition countries, we refer to all countries that were part of the Soviet Union plus the Central and Eastern European countries that were heavily influenced by the Soviet Union before 1990 (not including Albania and former Yugoslavia). Whenever we have been able to find the underlying data, we extend the GII measure back to the early 1990s. This extension allows us to measure the development of gender inequality through the lens of a single index since the beginning of the transition. We then discuss what the GII tells us about gender inequality in transition, but also – perhaps more importantly – what it does not tell us. Our analysis is discussed as well as shown in some more detail in our forthcoming companion FREE Policy Paper.

The Gender Inequality Index

The GII was reported for the first time in the 2010 Human Development Report. It measures gender inequalities in three dimensions of human development: 1) reproductive health, measured by maternal mortality and adolescent birth rates; 2) empowerment, measured by representation in parliament and secondary education among adults; and 3) economic status, measured by labor force participation.

GII country-values from 1995 are available on the UNDP website.  Conveniently for our purpose, most of the underlying data that the index is based on are also made available from the UNDP for the years 1990, 1995, 2000, 2005, and every year between 2010 and 2015, with the only exception of the female seat share in Parliament in 1990. Using the UNDP data, and data on the female seat share in Parliament in 1990 from additional sources (see the FREE Policy Paper for a list of sources), we obtain values for the GII from the beginning of the transition in 1990 until 2015.

What does the GII index tell us about gender equality in transition economies?

Figure 1 reports values for the GII index in box plots, which show the index 25th and 75th percentile (respectively bottom and top of the box), its median (horizontal line in the box), its maximum and minimum (whiskers), and outliers (dots) for two groups of countries: transition countries and Western-European countries. We have reconstructed the values of the GII index for a limited set of countries within these groups (see the note to Figure 1 for the list of countries). When interpreting Figure 1, recall that higher GII values imply more inequality.

Figure 1. The Gender Inequality Index in transition countries and Western Europe, 1990-2015

Nov122018_Figure1

Source: Own calculations based mainly on UNDP data. The transition countries are: Armenia, Bulgaria, Georgia, Hungary, Poland, Romania, and the Russian Federation. For Western Europe the countries are: Austria, Belgium, Cyprus, Denmark, Finland, France, Greece, Iceland, Italy, Luxembourg, Malta, the Netherlands, Norway, Portugal, Spain, Sweden, and Switzerland.

Figure 1 shows that based on the GII, median gender inequality is larger in transition countries than in Western Europe and has been so throughout the entire period since 1990. In both regions, the index shows a decreasing trend, after an initial increase in 1995 in the transition countries. As we show in the Policy Paper, this decrease is mainly due to a drop in female representation in national parliaments. The variance of the index scores has declined over time in Western Europe, while it remained mostly unchanged in the transition countries.

The evidence from the GII is somewhat at odds with the common notion that transition countries enjoy relatively low level of gender inequality. However, it is important to notice that transition and Western European countries are generally at different levels of development. Figure 2 displays the country groups’ performance in relation to their level of human development. This is done by measuring the difference between their GII ranking and their Human Development Index ranking (HDI) among all the countries with non-missing GII values in the years considered. The HDI is an UNDP-developed measure of overall human development. See the policy paper for details about its measurement. The larger the difference between GII- and HDI-ranking, the worse the group performance in terms of gender inequality in relation to its level of development.

Figure 2. Difference between Gender Inequality Index ranking and Human Development Index ranking in transition countries and Western Europe, 1990-2015

Nov122018_Figure2

Source: Own calculations based mainly on UNDP data.

The trends between transition countries and Western Europe are now opposite. In 1990, the median standing in terms of gender inequality was better than that in human development for transition countries, and the relative level of gender inequality was lower than in Western Europe. The (negative) difference between GII and HDI ranking however appears to have narrowed over time, and it is close to zero in 2015. Western European countries have instead improved their gender equality ranking in relation to their ranking in terms of human development over the period studied. Put differently, the ranking improvement in terms of human development in former socialist countries since the transition have not translated into comparable gains in gender equality ranking as measured by the GII index.

It is also important to emphasize that, according to several scholars, a dichotomy in terms of gender relations existed in transition countries during the socialist period. This is because on one hand the socialists put substantial into effort to empower women economically (see e.g. Brainerd, 2000; Pollert, 2003; Campa and Serafinelli, 2018), but on the other hand they failed to eliminate patriarchy (LaFont, 2001). This suggests that a composite index can mask important contrasting patterns among its components. In the Policy Paper we uncover such contrasting patterns. By looking separately at the different components of the GII index, we show that while Western European countries have invariantly improved their levels of gender equality since 1990, the trend in transition countries depends on the measure one looks at: Women maintained, but did not improve, their relative status in the labor force. They gained more equality in education and especially in terms of reproductive health, and lost descriptive political representation.

Conclusion

In this policy brief we have studied the development of gender inequality in transition countries through the lens of the Gender Inequality Index, whose span we have extended to the beginning of the transition period. We have shown that, based on this index, gender inequality has decreased since 1990 in transition countries, a trend which is common to that in Western Europe. However, once the changes in overall development during this period are taken into account, it appears that transition countries fared better in 1990 than today. Our analysis thus shows that analyzing gender inequality indexes in absolute terms and in relation to levels of development can deliver different conclusions. The factors that account for these differences should be kept in mind in policy discussions and policy-making. Some issues related to gender inequality, such as maternal mortality, are potentially addressed with a comprehensive strategy aimed at overall development. Conversely, other drivers of gender inequality, such as women’s political empowerment, do not necessary go hand in hand with overall development, and might therefore require more targeted policy interventions.

We have also cautioned the reader about the limitation of using comprehensive indexes to describe developments in gender inequality. A comprehensive index can overshadow important sources of gender inequality if it is composed of sub-indexes that move in opposite directions. This point can be especially relevant in the context of transition countries, which historically experienced a top-down approach to gender equality, the results of which in the long-term appear to be major advancements in some dimensions of women’s empowerment and contemporary potential backlash in other dimensions. It has been argued, for instance, that low levels of female representation in political institutions in transition countries can be the result of women’s large participation in the labor market while the division of roles in households remained traditional. In the words of anthropologist Suzanne LaFont (2001), “Women have been and continue to be overworked, and their lives have been over-politicized, the combination of which has led to apathy and/or the unwillingness to enter the male dominated sphere of politics. Many post-communist women view participation in politics as just one more burden”. In such a context, average values of an index of gender equality might mask high achievements in economic empowerment coexisting with lack of political representation.

References

  • Brainerd, E. (2000), ‘Women in Transition: Changes in Gender Wage Differentials in Eastern Europe and the Former Soviet Union’, Industrial and Labour Relations Review, 54 (1), pp. 138-162.
  • Campa, P. and Serafinelli, M. (2018), ’Politico-economic Regimes and Attitudes: Female Workers under State-socialism’, Review of Economics and Statistics, Forthcoming.
  • LaFont, Suzanne (2001), ‘One step forward, two steps back: women in the post-communist states.’ Communist and post-communist studies 34(2), pp. 203-220.
  • Pollert, A. (2003), ‘Women, work and equal opportunities in post-Communist transition’, Work, Employment and Society, Volume 17(2), pp. 331-357.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Energy Demand Management: Insights from Behavioral Economics

Posted on October 22, 2018 | Policy Brief

20181022 Energy Demand Management Image 01

It has long been recognized that consumers fail to choose the cheapest and most efficient energy-consuming investments due to a range of market and non-market failures. This has become known as the ‘Energy Efficiency Gap’.  However, there is currently a growing interest in terms of understanding on how consumers make decisions that involve an energy consumption component, and whether the efficiency of their decisions can be improved by changing the market incentives and governmental regulation. Meeting this interest, the most recent SITE Energy Talk was devoted to Demand Side Management.  SITE invited Eleanor Denny, Associate Professor of Economics at Trinity College Dublin, and Natalya Volchkova, Assistant Professor at the New Economic School (NES) in Moscwo and Policy Director at the Center for Economic and Financial Research  (CEFIR) to discuss the Demand Side Management process. The aim of this brief is to present the principles of Demand Side Management and discuss a few implemented programs in Europe, based on the discussions  during this  SITE Energy Talk.

For the last two decades, climate change policies have mostly been focused on the energy supply side, constantly encouraging new investments in renewables. But reducing energy demand may be as effective. Indeed, Denny and O’Malley (2010) found that investing 100MW in wind power is equivalent, in terms of emissions, to a decrease in demand of 50MW. Hence, there is a clear benefit of promoting energy saving. This has been the central point of different Demand Side Management (DSM) programs that may diversely focus on building management systems, demand response programs, dynamic pricing, energy storage systems, interruptible load programs and temporary use of renewable energy. The goal of these programs is to lower energy demand or, at least, smoothen the electricity demand over the day (i.e. remove peak-hour segments of demand to off-peak hours) as illustrated in Figure 1.

Figure 1 – Smoothing electricity demand during the day

A behavioral framework

DSM encompasses initiatives, technologies and installations that encourage energy users to optimize their consumption. However, the task does not seem easy, given the well-documented energy efficiency gap problem (e.g. Allcott & Greenstone, 2012 or Frederiks et al., 2015): consumers do not always choose the most energy efficient investments, despite potential monetary saving. One reason why might be that energy savings per se are not enough to trigger investment in energy efficient solutions or products. As Denny mentioned in her presentation, consumers will invest when the total  private benefits are higher than the costs of investment. This trade-off can be summarized by the following equation:

This equation illustrates that any DSM design should take into account both non-monetary benefits and consumers’ time preferences. The non-monetary benefits, such as improved comfort, construction and installation time, but also warm glow (i.e. positive feeling of doing something good) or social comparison, may play a major role. Moreover, the consumers’ time preferences (reflected here by the discount rate ) are also crucial in the adoption of energy efficient products. In particular, if consumers have present biased preferences, they would rather choose a product with a lower cost today and greater future cost than the reverse (i.e. higher cost today with lower future cost). Since energy-efficient products often require higher upfront investment, consumers that are impatient for immediate gains, may never choose energy efficient products.

Ultimately, it is an empirical (and context specific) question when and why DSM programs can reduce the energy efficiency gap. We describe below some DSM programs that have been implemented and discuss their impact.

Smart meters, a powerful DSM tool

A common DSM program is the installation of smart meters, which measure consumption and can automatically regulate it. The adoption of smart meters allows real-time consumption measures, unlike traditional meters that only permitted load profiling (i.e. periodic information of the customer’s electricity use).

Figure 2 – Energy Intensity in Europe

As illustrated in Figure 2, many European countries have implemented smart meter deployment programs. Interestingly, most of those countries have a relatively high level of energy efficiency (proxied by the energy intensity indicator of final energy consumption). On the contrary, in the Balkans and non-EU Eastern Europe countries, which fare poorly on the energy intensity performance scale, no smart meter rollout programs seem to be implemented.

Following the European Commission (EC) directive of 2009 (Directive 2009/72/EC), twenty-two EU members will have smart meter deployment programs for electricity and gas by 2020 (see Figure 2).  These programs are targeting end-users of energy, e.g. households that represent 29% of the current EU-28’s energy consumption, industries (36.9%) and services (29.8%) (EEA). With this rollout plan, a reduction of 9% in households’ annual energy consumption is expected.

The situation across the member states is however very different. Spain was one of the first EU countries to implement meters in 1988 for industries with demand over 5MW. All the meters will be changed at the end of 2018. 27 million euros for a 30-year investment in smart meter installations is forecasted (EC, 2013). Sweden started to implement smart meter rollout in 2003 and 5.2 million monthly-reading meters were installed by 2009. Vattenfall, one of the major utilities in Sweden, assessed their savings up to 12 euros per installed smart meter (Söderbom, 2012). Similarly in the United Kingdom, the Smart Metering Implementation Programme (SMIP) is estimated to bring an overall £7.2 billion (8.2 billion euros) net benefit over 20 years, mainly from energy saving (OFGEM, 2010). In general, smart metering has been effective, but its effectiveness may diminish over time (Carroll et al, 2014).

From smart-meter to real-time pricing

The idea of real-time pricing for electricity consumers is not new. Borenstein and Holland (2005) and Joskow and Tirole (2006) argue that this price scheme would lead to a more efficient allocation, with lower deadweight loss than under invariant pricing.

By providing detailed information about real-time consumption, smart meters enable energy producers to adopt dynamic pricing strategies. The increasing adoption of smart meters across Europe will likely increase the share of real-time-pricing consumers, as well as the efficiency gains. With the digitalization of the economy, it is likely that smart metering will grow. Indeed, Erdinc (2014) calculates that the economic impact of smart homes on in-home appliances could result in a 33% energy-bill reduction, due to differences in shift potential of appliances.

In 2004, the UK adopted a time-of-use programme called Economy 10, which provides lower tariffs during 10 hours of off-peak periods – split between night, afternoon and evening – for electrically charged and thermal storage heaters. The smart time-of-use tariffs involving daily variation in prices were only introduced in 2017.

Likewise, France’s main electricity provider EDF, implemented Tempo tariff for 350,000 residential customers and more than 100,000 small business customers. Based on a colour system to indicate whether or not the hour is a peak period, customers can automatically or manually monitor their consumption by controlling connection and disconnection of separate water and space-heating circuits. With this program, users reduced their electricity bills by 10% on average.

In Russia, the “consumptions threshold” program discussed by Natalya Volchkova, gave different prices for different consumption thresholds. But it seems that the consumers’ behaviour did not change. This might be due to the thresholds being too low, and an adjusted program should be launched in 2019.

Joskow and Tirole (2007), argue that an optimal electricity demand response program should include some rationing of price-insensitive consumers. Indeed, voluntary interruptible load programs have been launched, mainly targeting energy intensive industries that are consuming energy on a 24/7 basis. These programs consist of rewarding users financially to voluntarily be on standby. For instance, interruptible programmes in Italy apply a lump-sum compensation of 150,000 euros/MWh/year for 10 interruptions and 3000 euros/MW for each additional interruption (Torriti et al., 2010).

Nudging with energy labelling

Energy labelling has been also part of DSM. Since the EC Directives on Ecodesign and Energy Labelling (Directives 2009/125/EC and 2010/30/EU), energy-consuming products should be labelled according to their level of energy efficiency. For Ireland, Eleanor Denny has tested how labelling electrical in-home appliances may affect consumers’ decisions, like purchasing electrical appliances or buying a house. First, Denny and co-authors have nudged buyers of appliances, providing different information regarding future energy bills saving. They find that highly educated people, middle income and landlords are more likely to be concerned with energy-efficiency rates, rather than high-income people.

In another randomized control trial, Denny and co-authors manipulate information on the energy efficiency label for a housing purchase. In Ireland, landlords are charged for energy bills even when they rent out their property. The preliminary findings are that landlords informed about the annual energy cost of their houses are willing to pay 2,608 euros for a one step improvement in the letter rating – the EU label rating for buildings ranges from A to G – compared to the landlords that do not receive the information (see CONSEED project).

Similar to the European Directive, the 2009 Russian Energy efficiency law includes compulsory energy efficiency labels for some goods and improvements of the building standards (EBRD, 2011). Volchkova and co-authors run a randomized controlled experiment on the monetary incentives to buy energy efficient products. In 2016, people in the Moscow region received a voucher with randomly assigned discounts (-30%, -50% or -70%- for the purchase of LED bulbs. Vouchers were used very little, irrespective of the income. It seems that consumption habits and not so much monetary rewards were the main driver of LED bulb purchase.

How can DSM be improved?

Any demand response program requires some demand elasticity. For example, smart meters and dynamic pricing only improve electricity consumption efficiency if demand is price elastic. As Jessoe and Rapson (2014) show, one should provide detailed information (e.g. insights on non-price attributes, real-time feedback on in-home displays) to try to increase demand elasticity. Hence it seems that  the low adoption of energy efficient goods is partly due to a lack of information or biased information received by the consumers. First, it is difficult for many to translate energy savings in kWh in monetary terms. Second, many consumers focus on the short-term purchase cost and discount heavily the long run energy saving. These information inefficiencies can, in principle, be diminished by private actors and/or governmental regulation. Denny mentioned the possibility of displaying monetary benefits on labels in consumers’ decision-making in order to improve energy cost salience. For instance, in the US or Japan, the usage cost information is also displayed in monetary terms. Moreover, lifetime usage cost (i.e. cost of ownership) should also be given to the customers since it has been shown that displaying lifetime energy consumption information has significantly higher effect than presenting annual information  (Hutton & Wilkie 1980; Kaenzig 2010).

Summing up, DSM programs, including those with a behavioral framework, are an important tool for regulators, households and industries helping to meet emissions reduction targets, significantly decrease demand for energy and use energy more efficiently.

References

  • Allcott, Hunt ; Greenstone, Michael. 2012. “Is There an Energy Efficiency Gap?”, Journal of Economic Perspectives, 26 (1): 3-28.
  • Borenstein, Severin; Holland, Stephen. 2005. “On The Efficiency Of Competitive Electricity Markets With Time-Invariant Retail Prices”, Rand Journal of Economics, 36(3), 469-493.
  • Carroll, James; Lyons, Seán; Denny, Eleanor. 2014. “Reducing household electricity demand through smart metering: The role of improved information about energy saving,” Energy Economics, 45(C), 234-243.
  • Denny, Eleanor; O’Malley, Mark. 2010. “Base-load cycling on a system with significant wind penetration”, IEEE Transactions on Power Systems 2.25, 1088-1097.
  • Erdinc, Ozan. 2014. “Economic impacts of small-scale own generating and storage units, and electric vehicles under different demand response strategies for smart households”, Applied Energy, 126(C), 142-150.
  • European Bank for Reconstruction and Development. “The low carbon transition”. Chapter 3 Effective policies to induce mitigation (2011).
  • European Commission. Electricity Directive 2009/92. Annex I.
  • European Commission. Ecodesign and Energy Labelling Framework directives 2009/125/EC and 2010/30/EU.
  • European Commission. “From Smart Meters to Smart Consumers”, Promoting best practices in innovative smart metering services to the European regions (2013).
  • European Commission. “Benchmarking smart metering deployment in the EU-27 with a focus on electricity” (2014).
  • European Environment Agency. Data on Final energy consumption of electricity by sector and Energy intensity.
  • Frederiks, Elisha R.; Stenner, Karen; Hobman, Elizabeth V. 2015. “Household energy use: Applying behavioural economics to understand consumer decision-making and behaviour”, Renewable and Sustainable Energy Reviews, 41(C), 1385-1394.
  • Hutton, Bruce R.; Wilkie, William L. 1980. “Life Cycle Cost: A New Form of Consumer Information.” Journal of Consumer Research, 6(4), 349-60.
  • Jessoe, Katrina; Rapson, David. 2014. “Knowledge is (less) power: experimental evidence from residential energy use”, American Economic Review, 104(4), 1417-1438.
  • Joskow, Paul; Tirole, Jean. 2006. “Retail Electricity Competition“, Rand Journal of Economics, 37(4), 799-815.
  • Joskow, Paul; Tirole, Jean. 2007. “Reliability and Competitive Electricity Markets”, Rand Journal of Economics, 38(1), 60-84.
  • Kaenzig, Josef; Wüstenhagen, Rolf. 2010. “The Effect of Life Cycle Cost Information on Consumer Investment Decisions Regarding Eco‐Innovation”, Journal of Industrial Ecology, 14(1), 121-136.
  • OFGEM. “Smart Metering Implementation Programme” (2010).
  • Söderbom, J. “Smart Meter roll out experiences”, Vattenfall (2012).
  • Torriti, Jacopo; Hassan, Mohamed G.; Leach, Matthew. 2010. “Demand response experience in Europe: Policies, programmes and implementation”, Energy, 35(4), 1575-1583.

Project links

Eleanor Denny and co-authors’ European research projects:

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Focus on Investment: A Brief Look at Regulatory Developments in EU Telecommunications

Posted on April 15, 2018 | Policy Brief

20180415 Focus on Investment Image 01

The European Commission recently proposed a revision to its existing regulatory framework for telecommunications, the details of which have been amply discussed and are currently being negotiated. A pivotal theme of the revision is a stronger emphasis on stimulating investments into broadband networks capable of delivering high-speed (100+ Mbps) internet services. This brief highlights and briefly discusses some key changes in that regard.

Introduction

High-speed broadband networks are the backbone of the fast-growing digital economy. Promoting citizens’ access to such networks has been one of the European Commission’s stated policy priorities at least since 2010, when it launched its “Digital Agenda for Europe” (EC, 2014). Its policy mix of choice involves measures and funds facilitating deployment of so-called next-generation access networks on the one hand (commonly taken to mean access networks capable of delivering speeds exceeding 100 Mbps), while on the other hand regulating access to such networks to the extent perceived necessary to deal with potential problems resulting from incumbent network operators’ degree of market power. As regulation may harm incentives to invest in network infrastructure in the first place, a balance between investment promotion and competitive safeguards needs to be struck.

Motivated by what it considers to be a sub-optimally low speed of network upgrading in at least some of the EU’s member states, the Commission has sought to adjust its policy balance in favor of investments by proposing a revision (EC, 2016) of its regulatory framework for electronic communications, called the European Electronic Communications Code (EECC), which defines a standard approach to regulating fixed broadband network operators deemed to possess significant market power. That revision has been commented upon and discussed by the European Parliament and the European Council as well as various private and public stakeholders (Szczepański, 2017). Several amendments have been proposed and further discussion is ongoing to reach a compromise between the European institutions.

Background

Telecommunications networks were until more recently typically owned by vertically integrated, often formerly state-run, national incumbents who even after their privatization and the elimination of most legal barriers to entry were considered to possess significant market power. The EECC’s key remedy to such market power is so-called network unbundling at the wholesale level: considering the retail market for internet service provision potentially competitive, unbundling means granting competing internet service providers regulated access to the incumbent operator’s physical local-area access network, which is commonly regarded as the key bottleneck in internet service provision. Choosing the intrusiveness of the access obligation is up to the national regulatory authority (NRA), ranging from merely demanding that the incumbent publicly post a reference offer, to stricter measures such as non-discrimination, “fair and reasonable” pricing, and ultimately, full-on access price regulation, typically implemented with price caps derived from regulatory costing models. A recommendation from 2013 (EC, 2013) outlines methodological guidelines to national authorities.

Key changes

The proposed EECC revision makes the abovementioned recommendation binding, which may partly be an attempt to further harmonize regulatory practice between member states, with a view to encouraging cross-border investments by operators and service providers. It also encourages NRAs to, where possible, abandon more rigid price regulation in favor of margin squeeze tests. Margin squeeze occurs when a vertically integrated firm with market power in the wholesale segment of a production chain “squeezes” retail competitors by setting high wholesale and low retail prices, to the extent that even equally efficient, or at least reasonably efficient, retail competitors cannot survive if they are dependent on the dominant firm’s wholesale product. Moreover, and more importantly in terms of boosting deployment, the proposal encourages lighter-touch regulation for operators deploying new network infrastructure (Art. 72), and specifically relaxes regulation for deployment projects open to co-investments between operators (Art. 74). It also extends the market review period, i.e. the frequency at which NRAs are expected to update their market analysis and regulatory policy, from three to five years, giving operators a longer planning horizon, and encourages NRAs to consider any existing commercial wholesale offers in their market analysis, which can be interpreted to mean that anything short of full market foreclosure should be looked upon benevolently (Articles 61 and 65). In line with this latter development, which suggests a focus on wholesale access per se, is Article 77. This article exempts so-called wholesale-only networks – non-integrated networks whose very business model is selling access to interested internet service providers – from strict access price regulation, at least ex-ante. Typically, a presumption of consumer harm absent regulation is sufficient for intervention. Article 77 turns the tables on regulatory authorities by requiring evidence of actual consumer harm.

A counterpoint to these deregulatory elements is Article 59.2, which under certain conditions not only allows but obliges NRAs to impose access obligations on owners of existing physical infrastructure “up to the first concentration point”, in practice affecting mostly in-building wiring and cables, even when these owners have not been identified as dominant in any relevant market. In countries such as Sweden, where in-house wiring is often not owned by any operator but rather by the respective building’s owner(s), implementing such obligations may pose a regulatory challenge.

Finally, Article 22 requires NRAs to chart existing infrastructure as well as deployment plans across the country and enables them to define “digital exclusion areas” where no high-speed broadband infrastructure exists or is planned. In such areas, they may organize calls for interest to deploy networks, also with a view to resolving potential coordination problems between operators resulting from so-called “overbuild risk”: deployment in some lower-density areas may only be profitable if most of the customer base in that area can be captured, leading to a standoff between operators who cannot, do not want to, or are not allowed to communicate and coordinate their deployment strategies. As a result, investment is delayed.

A rather piquant detail here is that the proposed code allows NRAs to take action against operators it suspects of “deliberately” providing “misleading, erroneous or incomplete” information about their deployment plans. Included to prevent gaming, this provision carries the risk of suppressing investors’ appetite for the designated exclusion areas lest they be punished in case they change their mind. A minimum of mutual trust between the national regulator and market participants seems crucial for this mechanism to succeed.

Conclusion

The Commission’s proposed new regulatory framework emphasizes investment in, and take-up of, high-speed (100+ Mbps) broadband networks, explicitly defining such enhanced connectivity as a new regulatory objective on equal footing with the existing ones, most notably the promotion of competition. The present brief points out some key regulatory changes aimed at the fulfilment of these respective objectives. In terms of the revision’s impact on high-speed broadband deployment in the EU’s member states, it is difficult to make a general prediction since Europe is somewhat heterogeneous with respect to high-speed broadband penetration. For example, the 2016 EU overall NGA coverage was 75.9 % of households, but coverage rates of individual countries ranged from 99.95 % and 99.86 % in Malta and Belgium respectively to 47.0 % in France and a mere 44.2 % in Greece (EC, 2017). To the extent that the new code encourages investment relative to the old regime, regions with lower current coverage stand to benefit more. To the extent that the lower pace of deployment in those areas is the result of other factors orthogonal to regulation (one example being demand uncertainty), it will have a limited effect.

References

Paid Work after Retirement – Does Quality of Your Main Job in the Past Matter?

Posted on December 4, 2017 | Policy Brief

20171204 Paid Work after Retirement Image 01

In this brief, we summarize the results of a recent analysis focused on identifying the key determinants of engagement in paid work after retirement based on life histories data from the Survey of Health, Ageing and Retirement in Europe (SHARE). We find a strong link between the probability of work after retirement and indicators of quality of work prior to labor market exit, such as high physical and psychosocial demands, lack of control or receiving adequate social support. These results suggest a potentially important role of job-quality regulations. We find no significant association with past experience of adequate rewards with respect to efforts in the main job, which suggests that involvement in paid work after retirement may to a lesser extent be driven by financial concerns. This might mean that policy initiatives targeted at higher level of labor market activity among retirees should stress non-material aspects of employment in later life.

The collection of data in the 7th wave of the Survey of Health, Ageing and Retirement in Europe (SHARE) proceeded in 2017, and the Centre for Economic Analysis (CenEA) has recently published a report based on information collected in previous waves of the survey. The report entitled “The Polish 50+ generation in the European context: activity, health and wellbeing” examined among other issues the determinants of labor market activity of people aged 50+ with a special focus on Poland (Myck and Oczkowska, 2017).

SHARE is a panel survey conducted every two years and focuses on health conditions, material situation and social relations of the population aged 50 years and older. In 2017, in the 7th Wave, interviews were conducted with over 80,000 participants in 26 European countries and Israel. While the survey usually focuses on contemporary conditions of respondents, the interviews in Wave 3 (the SHARE-Life conducted in 2008-2009) is concerned with respondents’ life histories and topics such as family history, mobility and work histories.

In this brief, we draw on one of the chapters from the report and present results of a analysis that combines information on the quality of the main job of the respondents’ working careers, with information on engagement in paid work among retired individuals to examine key determinants of undertaking paid work after labor market exit.

Work histories in SHARE

The life-history interview includes a series of 12 questions evaluating effort-reward imbalance in the main job of individuals’ working careers (Siegrist and Wahrendorf, 2011; Siegrist et al., 2004; 2014). Based on these questions, five dimensions of the quality of the workplace were identified: physical and psychosocial demands, control, social support and reward (see Table 1). Figure 1 presents an example of the distribution of answers to one of the questions used to define these dimensions, which asked about the extent to which the respondents’ main jobs was physically demanding. Generally, men’s past main job is more often described as physically demanding than women’s. While less than half of respondents in France and Sweden claimed physically strenuous main job, the respective measure in Poland and Greece was as high as 75%.

Table 1. Dimensions of job quality

Dimension SHARE Questionnaire Items
Physical demands

– „My job was physically demanding.”

– „My immediate work environment was uncomfortable (for example, because of noise, heat, crowding).”
Psychosocial demands – „My work was emotionally demanding.”

– „I was exposed to recurrent conflicts and disturbances.”
Control – „I was under constant time pressure due to a heavy workload.”

– „I had very little freedom to decide how to do my work.”
Social support at work – „I received adequate support in difficult situations.”

– „There was a good atmosphere between me and my colleagues.”

– „In general, employees were treated fairly.”
Reward

– „I had an opportunity to develop new skills.”

– „I received the recognition I deserved for my work.”

– „Considering all my efforts and achievements, my salary was adequate.”

Notes: answer categories: “strongly agree, agree, disagree, strongly disagree”. Source: adapted from Siegrist and Wahrendorf (2011).

Figure 1. “My job was physically demanding”

Notes: includes wave 3 respondents with at least 10 years of seniority who retired by the time of wave 6; weighted. Source: own calculation based on SHARE data waves 3 (2008-2009) and 6 (2015).

Following Wahrendorf and Siegrist (2011), for the purpose of further analysis, we construct five measures of workplace quality based on the questions listed in Table 1. For each dimension of job quality, we calculate a sum-score of answers (from 1 “strongly agree” through 2 “agree”, 3 “disagree” to 4 “strongly disagree”) to selected questions, and identify the upper (lower) tertile of observations. We create five binary indicators (with 1 meaning “yes”) describing the quality of work in the sense of high physical or psychosocial demands, lack of control, and adequate social support or adequate reward. The results are presented in Figure 2 in association with the frequency of paid work after retirement.

Figure 2. Associations between quality of work in the past and frequency of paid work after retirement

Notes: includes wave 3 respondents with at least 10 years of seniority who retired by the time of wave 6 from selected countries (CZ, FR, DE, GR, PL, ES, SE); weighted. Source: own calculation based on SHARE data waves 3 (2008-2009) and 6 (2015).

In most cases the percentage of retirees engaged in paid work was significantly higher among those positively evaluating the quality of their past workplace. The only dimension where no significant difference was found in the level of involvement in paid work was between the retirees who estimated rewards at work as adequate to their efforts and those who assessed them otherwise.

What determines paid work after retirement?

The role of the five measures of job quality was further examined using models of probability of paid work after retirement. Apart from quality indicators regarding the main job, controls included total labor market experience, unemployment incidence, as well as detailed demographics and information concerning current health status and material conditions. Odds ratios were estimated separately for men and women from a group of selected countries: Czech Republic, France, Germany, Greece, Poland, Spain and Sweden.

Higher education is positively associated with the odds of employment after retirement, but have the opposite effect for age, poor health and living in rural areas. Each additional year of labor market experience increases the odds of working after retirement, but we find no significant effect of unemployment episodes.

Both men and women without experience of high physical demands and lack of control in their main job have higher odds of working after retirement than those who declared such experiences. For example, men who did not experience highly, physically demanding main jobs have 1.4 times higher odds of work after retirement compared to those who did. The respective odds for those who did not experience lack of control are 1.9. On the other hand, high psychosocial demands and adequate social support have significant influence only among retired women. Women who did not report high psychosocial demands had 1.25 times higher odds of work after retirement, while those who received adequate support in their past job had 1.5 times higher odds. We find no significant effect of the experience of adequate rewards with respect to efforts in the main job, and similarly no significant association between material conditions and employment of retirees. Both of these may imply that involvement in paid work after retirement is to a lesser extent driven by financial concerns.

Further discussion and policy implications

Differences in the degree of engagement in paid work after retirement with respect to the assessment of past job quality suggest a potentially important role of job quality regulations. At the same time, lack of significant association between the material situation and paid work after retirement implies that policy initiatives targeted at higher levels of labor market activity among retirees may benefit from stressing the non-material aspects of employment in later life.

Results point to a strong link between quality of work in the past and probability of work after retirement, which is in line with what other studies have showed: e.g. that low quality of work in the past strongly correlates with the desire to retire as soon as possible (e.g. Dal Bianco et al., 2014). Given the demographic pressure on public finances observed or expected in many developed countries, and foreseen reductions in the generosity of pension benefits, increasing the level of engagement in paid work after labor market exit may become an important policy challenge. The results summarized in this brief suggest that governments should, on the one hand, pay attention to the labor market conditions faced by those currently employed, and on the other hand focus on a broad set of incentives to encourage employment among older generations, going beyond financial remuneration.

References

Save

Save

Save

Save

Save

Save