Chinese evacuations and power projection (part 1): overseas citizen protection (The Strategist)

As China’s national and international economic interests have steadily grown, so has the People’s Liberation Army’s ability to protect them. China’s expanding, social-media-savvy middle class now expects the military to protect the country’s citizens overseas. That expectation has been reinforced formally by the strategic direction for the PLA to ‘protect the security of strategic SLOCs [sea lines of communication] and overseas interests’.

From a Chinese perspective, the increasing number of Chinese citizens living, working and travelling abroad has imposed an obligation on the state to protect them.

The Chinese leadership used to assume that, in most cases, other foreign powers would evacuate Chinese citizens along with their own from situations of unrest abroad. But things have changed over the past decade, and Beijing has introduced new policies and capabilities that have in turn generated new expectations among the Chinese population. More recently, President Xi Jinping’s Belt and Road Initiative has generated a further imperative to provide security abroad.

China’s approach to overseas citizen protection can be explained as the response of a rising power to the need to protect its people and interests overseas as its influence expands and its capability develops, acting as a responsible global power. Alternatively, the policy could be seen as a justification for the projection of power to underwrite China’s growing competitive stake in world affairs.

It’s important to note that Beijing uses the term ‘overseas Chinese’ to describe ‘Chinese citizens residing abroad and foreign citizens of Chinese descent’. That includes Han Chinese and any of the other ethnic groups from China, as well as people in other countries whose families haven’t been citizens of China for several generations. For example, 1.2 million Australian citizens identified themselves as having Chinese ancestry in the 2016 census, and only 41% of them were born in China. The use of ‘overseas Chinese’ as an official term, particularly in relation to China’s obligations to protect, clearly raises questions of sovereignty.

China’s 1982 constitution expresses the state’s intention to protect its people and interests abroad. But the Chinese Communist Party didn’t have the capacity to enforce its policy of ‘overseas citizen protection’ (海外公民保护) until recently. The catalyst for change occurred in 2004 when 14 Chinese workers were killed in Afghanistan and Pakistan. The domestic uproar compelled the CCP to acknowledge that large numbers of Chinese were living in high-risk environments overseas.

As a study by the Stockholm Peace Research Institute reveals, China conducted small-scale non-military operations to extract citizens in times of crisis or disaster from Solomon Islands, East Timor, Tonga and Lebanon in 2006, Chad and Thailand in 2008, and Haiti and Kyrgyzstan in 2010. Before that, the only evacuations of note had been from Indonesia in the mid-1960s and Kuwait in 1990.

In 2011 came the largest Chinese evacuations to date. During the Arab Spring uprisings, China repatriated 1,800 citizens from Egypt, 2,000 from Syria and 35,860 from Libya, along with 9,000 from Japan after the Tohoku earthquake.

The evacuation from Libya was the first to significantly involve the PLA, mainly in a coordination role. The operation took 12 days and involved 74 civilian aircraft, 14 ships, and around 100 buses.

After that experience, China’s military purchased more airlift and amphibious capability for situations in which civilian charters wouldn’t be available, such as in remote locations or high-threat environments. While that equipment can be required for evacuation operations, it is also fundamental to force projection.

The evacuation from Yemen in 2015, carried out exclusively by the military, was the first time the Chinese navy evacuated citizens of other countries as well. Between 30 March and 2 April, the naval command diverted a flotilla of three PLA Navy vessels from a counter-piracy mission in the Gulf of Aden to the port of Aden.

They evacuated 629 Chinese nationals and 279 citizens from 15 other countries (including Germany, India and the UK) to Djibouti. Chinese observers viewed the operation as a successful demonstration of the navy’s new rapid-reaction capabilities, developed through missions in the Gulf of Aden and off the Somali coast.

Beijing continues to face a rising expectation to protect its people overseas, as shown by the public reaction to the killing of Chinese citizens by militants in Syria, Mali and Pakistan. Other evacuations have included the extraction of Chinese workers from Samarra in Iraq in 2014, Chinese embassy and medical staff from Sudan (after two Chinese peacekeepers were killed) in 2016, and Chinese citizens from Dominica, Antigua and Barbuda in the Caribbean after a hurricane. Last year, China evacuated more than 2,700 tourists from Bali after a volcanic eruption.

The Central Military Commission decided in 2018 to expand the PLA Marine Corps from 20,000 to 100,000, with some units to be stationed overseas. One source claims 10,000 of these marines will be based at China’s first declared overseas military base, in the port of Obock in Djibouti, and more are likely to be based at Gwadar in Pakistan. They will protect Chinese oil imports from the Middle East at strategic nodes along the ‘belt and road’.

The new marines and bases are part of China’s strategy of ‘far seas protection’ (远海护卫), which requires the navy to safeguard overseas interests, including resources, strategic sea lanes, overseas citizens, investments and commercial entities. It has led to the development of a blue-water navy capable of operating globally with aircraft carriers and amphibious capabilities.

As the PLA’s capability to project force and protect Chinese citizens overseas continues to grow, so does the CCP’s readiness to implement the policy of overseas citizen protection, and the Chinese people’s expectation that it will do so.

As China’s global influence expands, unexpected friction could arise when the policy of overseas citizen protection is applied to situations outside China’s traditional core interests. Such actions are at odds with Beijing’s frequently enunciated foreign policy principle of ‘non-interference’.

One hypothetical example would be a recurrence of the anti-Chinese riots experienced in Tonga and Solomon Islands in 2006 and Papua New Guinea in 2009. Significant numbers of Chinese citizens could be at risk in Melanesia, where traditional partners (such as Australia, New Zealand and the United States) expect to provide the forces for evacuation and, if necessary, stabilisation operations.

Without adequate preparation, clear communication and shared understanding, an evacuation operation by the PLA in the South Pacific could lead to unintended consequences, raising questions of sovereignty, citizenship and jurisdiction. Such a development could be disadvantageous to the host nation, its traditional partners and China. This concept has been described elsewhere as ‘accidental friction’.

In part 2, I will explore recent trends in Chinese filmmaking that aim to generate patriotic pride in domestic audiences, while also attempting to normalise the perception of Chinese power projection within the international community.

Counterterrorism cooperation in the Maghreb: Morocco looks beyond Marrakech (The Strategist)

Seven years after its last major terrorist attack, Morocco’s multifaceted strategy for dealing with Islamist extremism appears to be paying off security-wise—but at the expense of liberal reforms.

Morocco lies in an unstable region with an ongoing Islamist terrorism problem but, with some international help, it seems to have regained control and hasn’t had an internal terrorist attack since April 2011. The kingdom portrays the fight against Islamist extremism as a war of ideas that can be won by preaching a tolerant, moderate and non-violent form of Islam.

Before that, there were several terrorist attacks. In May 2003, suicide bombers carried out the deadliest terrorist attack in Morocco’s history, in Casablanca—45 people died, including the 12 suicide bombers who came from the nearby shanty town of Sidi Moumen. In 2007, several suicide-bomb attacks in Casablanca killed the perpetrators and one other person. In April 2011, 17 people were killed in Marrakech when a bomb concealed in a bag was detonated at the Argana café, a popular tourist spot; the dead included a group of French students.

After the 2003 Casablanca attack, Morocco implemented a counterterrorism strategy based on a three-pillar system: strengthening internal security, fighting poverty, and undertaking religious reforms.

It also stepped up international security cooperation.

Assistance from the US under its Antiterrorism Assistance program ensured that Morocco and its neighbours have common methodologies for countering terrorism.

Since 2001, the Western Mediterranean Forum, also known as the ‘5 + 5 Dialogue’, an informal political, economic and cultural grouping of five countries from the Maghreb region (including Morocco) and five from southern Europe, has been a mechanism for addressing common terrorism concerns.

In 2004, 14 regional countries, including Morocco, established the Middle East and North Africa Financial Action Task Force to focus on the threat of terrorism-financing activities across the region.

In 2005, the US initiated the Trans-Sahara Counterterrorism Partnership to help 11 West African and North African countries (including Morocco) counter radicalisation and boost their counterterrorism capabilities.

Morocco hasn’t taken the lead on any of these counterterrorism initiatives, but since 2011 it has become more proactive.

One example is its involvement in the Global Counterterrorism Forum, an international forum of 30 member states, including Australia, established in 2001. Morocco was a founding member and in 2016 became co-chair of the forum’s coordinating committee.

In 2013, Morocco, together with the US, announced the Initiative to Foster Cooperation Networks among Justice Sector and Other Law Enforcement Practitioners in the Sahel and Maghreb Region, with the aim of facilitating cross-border investigations.

Morocco and the US also promoted the Border Security Initiative, adopted in 2016 by the Global Counterterrorism Forum and the United Nations Counter-Terrorism Centre to secure porous national borders through cross-border cooperation.

Morocco and Algeria are the most influential countries in the Maghreb, but their ongoing border dispute has at times been an impediment to regional cooperation on counterterrorism.

The conflict in Syria and Iraq has also had significant repercussions for Morocco. Between 2014 and 2016, some 1,600 young Moroccan men and women travelled to join Islamic State in Syria. Return of veteran extremists continues to be a major security concern.

In response to the Islamic State challenge, Morocco adopted three major national counterterrorism measures.

The first was a ‘home affairs’ approach to security. In October 2014, the Ministry of the Interior launched Operation Hadar (Arabic for ‘vigilance’), requiring the security forces to work together to combat terrorism. This resulted in enhanced security at airports, train stations and border posts.

The second measure was to strengthen the legal system. In 2015, the Moroccan parliament amended the country’s counterterrorism laws to criminalise a range of possible terrorism-related activities, including travel to Syria.

The third measure was establishment of the Central Bureau of Judicial Investigation as the lead agency for counterterrorism in Morocco. Since 2015, it has been responsible for numerous terrorism-related arrests and is credited with preventing several terrorist attacks in Morocco.

European countries have also worked closely with Morocco, providing counterterrorism training and equipment, and conducting joint operations with Moroccan security forces.

Morocco is said to be a capable security partner that closely monitors its population and controls domestic religious activities. But to do so, it relies on a repressive political and security system. The powerful Moroccan security service, the General Directorate of Territorial Surveillance, is believed to abduct and torture regime ‘enemies’—including political activists, suspected terrorists and separatist dissidents. Morocco continues to resist outside calls for liberal reform.

Australia now has an embassy in the capital, Rabat. Former foreign minister Julie Bishop announced the embassy’s opening on 16 November 2016, describing Morocco as ‘an important addition to Australia’s diplomatic network in Africa’.

Although Australia participates in international counterterrorism forums with Morocco, because of its distance from our region, we have no bilateral counterterrorism relationship with Morocco—nor do we have any official counterterrorism involvement in the Maghreb.

The regional factors bringing Turkey and Iran together (The Strategist)

US President Donald Trump’s policy of putting economic pressure on Iran to force regime change by inciting a domestic revolt seems to be failing. There is little doubt that renewed sanctions have hurt Iran economically, as witnessed above all by the precipitate fall of the Iranian currency in their wake. However, economic pressure has not led to a revolt against the regime and the currency has stabilised after the initial shock. Indeed, the American action may have consolidated support behind the regime, which can now deflect criticism of its economic performance on to the imposition of American sanctions.

In this context, Iran’s relations with Turkey provide a very interesting case study. Both Tehran and Ankara have regional ambitions that have sometimes led to friction between them, as was the case over Syria until recently. However, economic complementarities and congruence of strategic interests have helped to keep their relationship on a relatively even keel.

When the Trump administration announced that it was going to reimpose sanctions on Iran, Turkey made it clear that it wouldn’t follow American diktats but would comply only with sanctions imposed by the UN. Economic interdependence provides part of the explanation for the Turkish stand. Bilateral trade between Iran and Turkey isn’t limited to oil and gas. The volume of trade between the two neighbours stood at US$11.7 billion at the end of 2017, up from US$9.7 billion in 2016, and both countries have committed to eventually raising the level to US$30 billion.

However, it’s not just oil and trade that determine Turkish–Iranian relations; there’s also a convergence of political objectives. Turkish and Iranian strategic interests coincide on Kurdish secessionism, which threatens the territorial integrity of both countries. That’s why Iran didn’t oppose Turkish incursions into Syria to prevent the creation of a Kurdish enclave abutting its borders, even when the two countries supported opposite sides during the civil war. Now that Turkey is reconciled to Iran’s ally Bashar al-Assad remaining in power in Syria, the major political disagreement between Ankara and Tehran has lost its importance.

Iran’s support to Turkey’s president Recep Tayyip Erdoğan at the time of the failed coup in July 2016 greatly helped in patching up differences. The Iranian foreign minister stayed up all night as the coup was unfolding to monitor the Turkish situation and telephoned his Turkish counterpart five times to express Iran’s support for the government, thus strengthening personal bonds between the leaders of the two countries.

There’s also an increasing conjunction of interests between the two countries vis-à-vis Saudi Arabia. Tehran has been engaged for years in a fierce competition with Riyadh over primacy in the Gulf and over their respective roles in the wider Middle East. Syria had been the primary battleground for their rivalry since 2011. Now that the Syrian civil war is almost over, Yemen has become the major arena of conflict between them. Saudi Arabia and its ally the United Arab Emirates are engaged in open warfare with the Houthis who are in control of the Yemeni capital and are supported by Iran. The Saudi–UAE aerial bombardments have ravaged an already desperately poor country, killing thousands of civilians. An estimated eight million people are on the verge of starvation.

Ankara has increasingly come to see Riyadh as its primary antagonist in the competition for influence in the Sunni countries of the Middle East. It finds Tehran a useful ally in tying down Saudi Arabia in the Persian Gulf, thus making it easier for Turkey to emerge as the preeminent Sunni power in the rest of the region.

Saudi Arabia’s imposition of an economic blockade on Qatar further strained relations between the two countries. Turkey has a military base in Qatar, which it reinforced following the imposition of the blockade, and the emir sent a contingent of Qatari troops to provide security to the Turkish president at the time of the coup in 2016. Turkey reciprocated in 2017 by flying in essential commodities to Qatar in tandem with Iran in order to render the Saudi blockade redundant. The convergence of interests on Qatar between Tehran and Ankara arises from the primary reason for the Saudi blockade: Qatar’s cordial relations with neighbouring Iran, with which it shares the world’s largest gas field.

Jamal Khashoggi’s murder in the Saudi consulate in Istanbul has greatly infuriated the Turkish government, especially the president, who has taken it as a personal insult. Erdoğan has asked for the extradition of Khashoggi’s murderers to Turkey to stand trial and has clearly implicated Saudi crown prince Mohammed bin Salman in the crime, without naming him directly. Turkish–Saudi relations are currently at the lowest point in their history. This has worked to Iran’s advantage and has further cemented relations between Tehran and Ankara.

Turkey and Iran are thus moving towards a joint front against Saudi Arabia and its allies, whose numbers are dwindling as the Gulf Cooperation Council unravels. The United States would be wise to take this configuration of forces into account while formulating its future policy on the Middle East.

Curbing radical sermons in Indonesia (East Asia Forum)

Indonesia’s national intelligence agency (BIN) revealed in November 2018 that 41 mosques connected to government institutions and state-owned enterprises conduct sermons laced with extremist advocacy. The discovery was part of a survey by the Nahdlatul Ulama Islamic Boarding School and Community Development Association (P3M NU), which identified as many as 500 mosques suspected of radicalising worshippers.


The advancement of radical messages in formal institutions in Indonesia is not new. Several studies have uncovered support for subversive Islamist organisations and militant preaching within state university mosques in recent years. The Indonesian government is falling short in addressing the problem on campuses, and the new revelations may further test its willingness and ability to intervene.

Faith-based intolerance in Indonesia has been in the spotlight since the bitter campaign for the April 2017 gubernatorial elections in the special administrative region of Jakarta. The incumbent governor Basuki Tjahaja Purnama, commonly known as Ahok, was accused and later prosecuted for a comment he made during a campaign speech that was deemed blasphemous.

When anger began to swirl following the widespread circulation of the offending speech, fiery rhetoric increasingly imbued mosque sermons throughout the country. Once Ahok received a prison sentence for his words, the fanaticism did not appear to simmer down.

In an attempt to counter this wave of toxic discourse, the Ministry of Religious Affairs released a list of 200 preachers in May 2018 who they accredit with having deep knowledge of Islam and adhering to the founding values of the Indonesian Republic. Minister Lukman Hakim Saifuddin stressed that the initial list was merely a first step toward a larger database of endorsements, and that the ministry will work together with the Indonesia Ulama Council and Islamic community organisations to perfect the system.

The ministry also re-emphasised the obligation for preachers to read government-issued sermon guidebooks, especially the sections on love for one’s country and national identity. The general public was also encouraged to report any subversive or intolerant messages being spread in local mosques.

Criticism of the state-sanctioned preacher list was swift. A number of lawmakers and religious scholars asked if the directory meant that anyone who didn’t make the cut was either incompetent or extremist. Among the sceptics was Indonesian Vice President and Chair of the Indonesian Mosque Council (DMI) Jusuf Kalla, who recently convened a meeting with the DMI board to draw up plans for devising a new set of guidelines for preachers.

For their part, the Indonesian National Police (Polri) followed up on the P3M NU report by mapping and profiling the identified mosques, and by committing to cooperation with relevant agencies such as the Ministry of Religious Affairs and local governments. The Polri are also conducting a campaign to raise awareness among communities about the dangers of extremism and the possibility of legal action being taken against preachers who are proven to deliver sermons with radical messages.

Yet some politicians question the report’s veracity. Members from the National Mandate Party (PAN) and the Prosperous Justice Party (PKS) argue that BIN has overestimated the problem and blown it out of proportion. Members of the House of Representatives and the Expert Council of the Association of Indonesian Muslim Intellectuals suggest that because the information submitted by BIN was a PBNU finding, it could harbour political intentions aimed at cornering certain parties — particularly in the lead up to general elections in April 2019.

This politicisation of BIN’s revelations is unfortunate and potentially dangerous. Indonesia has been experiencing a notable rise in levels of publicly expressed intolerance in recent years — from ostracism of religious minorities and violence toward LGBT communities to strict interpretations of correct religious practice and intense scrutiny on adherence. The nation has also witnessed an upsurge in violence from those who support foreign terrorist organisations and seek to establish a caliphate in Southeast Asia.

Downplaying the sobering discovery that several hundred mosques may be proliferating extremist arguments is most likely itself a political move to appeal to the religiously conservative bases of PAN and PKS. It may even represent a broader strategy to adapt existing norms toward greater accommodation of hard-line religious views and exclusivism.

Government officials at both central and regional levels should keep in mind the principles of Indonesia’s national motto Bhinneka Tunggal Ika (unity in diversity) when considering how societal values may be evolving. As a pluralistic country with a diverse array of ethnic groups and six official religions, state institutions need to ensure harmony by ironing out threats to national unity while maintaining freedom of speech.

Recent initiatives suggest that some government officials and agencies may be serious about reining in those who preach bigotry and contempt. Time will tell whether such efforts can weather the resistance.

Yuslikha K Wardhani is a Researcher at the Research Center of Police Science and Terrorism Studies, School of Strategic and Global studies, University of Indonesia.

Cameron Sumpter is an Associate Research Fellow at the Centre of Excellence for National Security, a constituent unit of the S Rajaratnam School of International Studies, Nanyang Technological University, Singapore.

The War on Huawei (Project-Syndicate)

The Trump administration’s conflict with China has little to do with US external imbalances, closed Chinese markets, or even China’s alleged theft of intellectual property. It has everything to do with containing China by limiting its access to foreign markets, advanced technologies, global banking services, and perhaps even US universities.


Ending Wartime Sexual Violence (Project-Syndicate)

This year’s Nobel Peace Prize has been awarded to two activists for their efforts to end the use of sexual violence in armed conflict. But if their efforts are to lead to real change, heightened awareness must be turned into concerted action, and recent research can help.


The micro impact of macroprudential policies: Firm-level evidence (VOX)

Macroprudential policies have been the focus of increased attention in the post-crisis regulatory reform agenda. These policies are intended to limit systemic risk – the risk of disruptions to the provision of financial services caused by impairment to parts or all of the financial system (IMF 2013).

While an expanding literature has explored the effect of macroprudential policies on the aggregate economy and bank-level credit (Akinci and Olmstead-Rumsey 2017, Cerutti et al. 2015, Claessens et al. 2013), there has been little empirical evidence on the relationship between macroprudential policies and firm outcomes. Importantly, macroprudential policies might have differentiated effects across different types of firms. In particular, evidence that the ex-ante most financially constrained firms tend to be most affected by macroprudential policies could be suggestive of a trade-off between financial stability and inclusion.

Our recent research (Ayyagari et al. 2018) assesses the effectiveness of macroprudential policies and their impact on firms Specifically, using data across 900,000 firms from 49 countries for the period 2003-2011, we gauge the differential impact of macroprudential policies on small and young firms that tend to be more financially constrained to begin with and that the literature has found are more responsive to policy shocks. Our findings are important for policy formulation both in terms of the effectiveness of macroprudential policies and in terms of their unintended consequences.

Why macroprudential tools?

The case for macroprudential policies rests on the notion that a high correlation in the behaviour across financial institutions can result in contagion effects which can cause idiosyncratic distress to become systemic. In addition, strong credit cycles may have the potential not only to exacerbate business cycles, but also to lead to systemic banking distress. In the broadest sense, one can distinguish between a cross-sectional and a time-series dimension of macroprudential tools. In the cross-sectional dimension, macroprudential policies (e.g. higher capital requirements or regulatory restrictions on institutions whose failure would have a strong negative impact on the system) seek to limit the build-up of vulnerabilities that arise through linkages across financial institutions and from the critical role played by some institutions. In the time-series dimension, macroprudential policies intend to reduce systemic vulnerabilities arising from procyclical feedback between asset prices and credit and limiting unsustainable increase in leverage and volatile funding (IMF 2013).  Our empirical analysis focuses on on the use of macroprudential tools to smooth credit cycles over time.

We distinguish between: (i) tools targeted at borrowers’ leverage and financial positions, and (ii) tools targeted at financial institutions using data from the Global Macroprudential Policy Instruments, as described inCerutti et al. (2015). The former includes loan-to-value and debt-service-to-income ratios, while the latter includes the following ten instruments: dynamic loan-loss provisioning, countercyclical capital buffers, bank leverage ratios, capital surcharge for systemically important financial institutions, limits on interbank exposures, concentration limits, limits on foreign currency loans, limits on domestic currency loans, reserve requirement ratios, and taxes or levies on financial institutions. We further use recent data on the intensity of macroprudential tools (Cerutti et al. 2017), as discussed below.

Data and methodology

To explore the relationship between macroprudential policies and firms’ financing, investment and sales growth, we combine the macroprudential indicators with data from Orbis, a commercial database distributed by Bureau van Dijk which contains basic firm-level information including data on external financing for over 900,000 companies across 49 countries over the period of 2003 to 2011. Compared with other databases, the unique advantage of using Orbis is that it includes data on large, small, listed and unlisted firms.  We explore both short-term (with residual maturity of less than one year) and long-term (with residual maturity of one year of more) financing, as well as investment and sales growth.

There are several advantages to using micro data to examine the impact of macroprudential policies.  First, using firm-level data and focusing on the differential effects of macroprudential policies across firm groups helps to mitigate endogeneity concerns regarding the adoption of macroprudential policies, as it is harder to argue that credit developments in individual firms or specific firm groups will drive the adoption of aggregate macroprudential policies. Second, by conducting the analysis at the firm level we can include country and time fixed effects to control for the impact of other macroeconomic developments (e.g. monetary policy) that might also affect firm credit growth.

We run firm-level regressions of financing, investment and sales growth on country-year indicators of macroprudential policies and control for other macroeconomic factors. We also focus on the differential relationship between macroprudential policies, financing, investment and sales growth of small and young firms relative to other firms. This allows us to control for any confounding time-varying country factors that might affect financing growth for any average firm in the country.

Our findings

Our results suggest a significant association between the implementation of some macroprudential tools and firms’ financing and real sector growth, but we also find evidence of differential effects across different firm groups.

  • There is a significant and negative association between borrower-targeted macroprudential policies and firms’ long-term financing growth. Specifically, applying one additional borrower-related macroprudential tool is associated with 4.8% lower long-term financing growth, while the correlation between financial-institution targeted macroprudential tools and firms’ financing growth is not significant.
  • Micro, small and medium-sized firms’ as well young firms’ growth in short- and long-term financing  decreases with the additional implementation of borrower-related macroprudential tools, while the association between financial institution-targeted macroprudential tools and financing growth is relatively stronger and more negative for micro firms.
  • Both the implementation and intensity of macroprudential tools are significantly associated with firms’ financing growth. Both the level and the change in the loan-to-value ratio are associated with a relatively lower long-term financing growth of micro, small and medium sized enterprises, while neither short-term financing growth nor overall financing growth seem to be impacted. These findings hold both for microenterprises and small and medium enterprises, but not for young firms.
  • Less financially healthy firms are more affected by the implementation of macroprudential policies.  Specifically, highly levered and less profitable firms and firms with low interest coverage experience stronger reductions in financing growth than other firms after the implementation of macroprudential policies.
  • Macroprudential policies are also associated with real sector outcomes.  Specifically, applying one additional borrower-targeted macroprudential tool is associated with a 4.4% lower investment growth and 3.5% lower sales growth. As before, there is no significant association with financial institution-targeted macroprudential tools. The link between macroprudential tools and investment and sales growth is stronger for micro, small and medium-sized enterprises and for younger firms.

Concluding remarks

Our results show that macroprudential tools affect firms’ financing, investment and sales growth, speaking for their effectiveness. However, it is borrower-targeted policies rather than measures targeted at banks that are most effective, consistent with previous findings that macroprudential measures targeted at banks are subject to leakage (Aiyar et al. 2014).

The effectiveness of macroprudential tools works primarily through reducing financing growth for micro, small and medium-sized enterprises and young firms that have fewer alternative financing sources. However, it is the financially less healthy firms among these groups that experience a higher reduction in financing growth, thus providing limited evidence for a trade-off between financial stability and inclusion.

This finding is consistent with the literature that has found that small firms are more affected by policy changes (Forbes 2007, Gertler and Gilchrist 1994). Furthermoreit points to a trade-off between financial stability and financial inclusion. Reassuringly, however, banks seem to be reducing financing growth primarily to riskier (more financially fragile) firms.

Authors’ note: The views expressed here are those of the authors and do not necessarily represent those of the institutions with which they are affiliated.


Aiyar, S, C W Calomiris and T Wieladek (2014), “Does macroprudential regulation leak? Evidence from a UK policy experiment”, Journal of Money, Credit, and Banking 46(1): 181-214.

Akinci, O and J Olmstead-Rumsey (2017), “How effective are macroprudential policies? An empirical investigation”, Journal of Financial Intermediation 33: 33-57.

Ayyagari, M, T Beck and M S Martinez Peria (2018), “The micro impact of macroprudential policies: Firm-level evidence”, IMF working paper 18/267.

Cerutti, E, S Claessens and L Laeven (2015), “The use and effectiveness of macroprudential policies: New evidence”, IMF working paper 15/61.

Cerutti, E, R Correa, E Fiorentino and E Segalla (2017), “Changes in prudential policy instruments: A new cross-country database”, International Journal of Central Banking 13(1): 477-503.

Claessens, S, S Ghosh and R. Mihet (2013), “Macroprudential policies to mitigate financial system vulnerabilities”, Journal of International Money and Finance 39: 153-185.

Forbes, K (2007). “One cost of the Chilean capital controls: Increased financial constraints for smaller traded firms”,  Journal of International Economics 71(2): 294-323.

Gertler, M and S Gilchrist (1994), “Monetary policy, business cycles, and the behaviorof small manufacturing firms”, Quarterly Journal of Economics 109(2): 309-340.

IMF (2013), “Key aspects of macroprudential policy”, IMF policy paper.

Meghana Ayyagari, Thorsten Beck, Maria Soledad Martinez Peria

Regulatory framework for the lending-based crowdfunding platforms (VOX)

Lending-based crowdfunding platforms represent a new mode of financial intermediation by connecting lenders and borrowers directly using internet platforms. While their share of the loan market is still very small, in some niches these platforms are a real alternative to bank credit. For example, in 2017, business lending by UK crowdfunding platforms amounted to almost 30% of new loans to small businesses (Zhang et al. 2018) (see Figure 1).

Figure 1 Business lending-based crowdfunding in the UK

Source: Zhang et al. (2018).

Technically, lending-based crowdfunding platforms do not perform risk and maturity transformation, but they are experimenting with business models that could allow them to perform bank-like functions in the future (Havrylchyk and Verdier 2018).

  • Many platforms allow lenders to automate their lending process by setting lending criteria (risk, maturity, and so on). This lowers transaction costs and permits diversification.
  • Platforms use credit scoring to assign a risk band to every borrower and effectively play the role of a delegated monitor insofar as lenders delegate to them due-diligence.
  • Some platforms provide liquidity services by creating secondary markets in which lenders sell their loans to other investors.

Regulation of crowdfunding

To review existing regulatory practices, we sent a questionnaire about regulation of crowdfunding to all OECD countries, and received 17 replies (Havrylchyk 2018). While many OECD countries require lending-based crowdfunding platforms to adapt to existing regulation, nine of the responding countries had put in place legislation designed to regulate this mode of financial intermediation.

In March 2018, following consultations, the European Commission presented a proposal for an EU-wide passporting regime for both lending-based and investment-based crowdfunding (European Commission 2018).

Table 1 Summary of activities that are allowed by lending-based crowdfunding platforms

Notes: Yes/No – if activity is allowed/not allowed; NR – if activity is not regulated or not mentioned in the law. Information is provided separately for two legal statuses of lending-based platforms that facilitate loan agreements and investment-based platforms that facilitate debt securities.
Source: OECD Questionnaire in Havrylchyk (2018).

Table 1 summarises the results of the questionnaire for countries that have put in place specifically designed regulation for lending-based crowdfunding. These regulatory texts do not propose any instruments, but rather define the scope of activities of crowdfunding platforms. While some regulators restrict crowdfunding platforms to simple credit intermediation, others explicitly set high maximum amounts for originated loans, allow automated lending, provision funds, and secondary markets, and permit a platform to invest in loans that it facilitates.

Restricting crowdfunding platforms to credit intermediation limits short-term risks, but it also prevents platforms from experimenting with business models, and eventually competing with banks. In this context, the recent European Commission regulatory proposal appears to be conservative, since it does not mention automated lending, limits the maximum loan amount to €1 million, explicitly forbids platform from investing in loans, and limits the responsibility of platforms in the organisation of secondary markets (called ‘bulletin boards’).

There are three types of market failures in the financial system that need to be addressed by the regulators: coordination problems and runs, moral hazard and adverse selection, and market power or barriers to entry.

Coordination problems and runs

Large lending-based crowdfunding platforms offer secondary markets, which are a useful feature for providing liquidity to investors and could be essential if platforms are to scale up. Theoretically, platforms with secondary markets would not be subject to coordination problems and self-fulfilling runs in the same way as banks are. The direct link between lenders and borrowers ensures that lenders’ strategy (to run or not to run) does not depend on the strategy of other lenders, but only on the solvency of borrowers in their loan portfolio.

Nevertheless, secondary markets might become illiquid and misprice traded loans. Also, it would be misleading for platforms to promise lenders maturity transformation (that is, their investment can be always liquidated). Currently, secondary markets are forbidden in many countries – and in some cases the regulatory approach is not clear, as Table 1 shows. In particular, the EU proposal explicitly states that platforms cannot put in place a trading system but, instead, proposes a bulletin board that allows investors to interact directly with each other to buy and sell loan agreements or transferable securities. Importantly, this buying and selling activity on crowdfunding platforms is at the client’s own discretion and responsibility.

Regulatory challenges: Coordination problems and runs: 

  • How do we ensure well-functioning secondary markets?
  • How can we ensure that platforms do not promise the transformation of maturity to lenders?

Adverse selection and moral hazard

Retail lenders face severe adverse selection problems when choosing to whom they lend. This problem is particularly acute because most regulators set a maximum loan size (Table 1). This allows lending only to small businesses, and this type of lending tends to be riskier and opaque. To mitigate this adverse selection, lenders delegate due diligence to crowdfunding platforms. Hence, it is important to ensure that platforms have good risk management systems. Platforms are likely to experiment with new methods of credit scoring that rely on big data and machine learning. While these techniques are promising, they are untested. Importantly, due diligence and scoring models are currently not supervised.

Since lenders bear all credit losses, aligning incentives between platforms and lenders should be at the heart of regulation. To signal that they retain some risk, several platforms invest in loans that they facilitate, while others use their own capital to create provision funds that absorb first losses. Indeed, if platforms do not retain the risk of the loans they facilitate, their business model could resemble the ‘originate and distribute’ model of the securitisation process. This was one of the major causes of the global crisis, as it loosened credit standards.

In the EU proposal, platforms are explicitly forbidden to invest in loans that they facilitate in several countries (Table 1). This rule is motivated by the potential conflict of interest, because platforms might cherry-pick the best loans. this is a legitimate concern. If platforms invest in any loans, they should invest in allloans. A well-designed minimum capital requirement could be used as a tool to provide incentives to platforms managers and shareholders not to expose lenders to excessive risks.

Finally, if platforms become systemically important in the future, it is important to ensure that an orderly resolution is possible – in other words, the continuation of loan repayments so that lenders do not lose money if the platform fails. This is important because if there are financial institutions that are too big or too interconnected to fail, this increases the likelihood of state bail-outs. This intensifies moral hazard problems. No large platform has yet failed, but the direct connection between lenders and borrowers means it would be easier to design an orderly resolution for a platform than for a bank. Since all lenders bear their own losses, it is comparable to a bail-in mechanism.

Regulatory challenges: Adverse selection and moral hazard 

  • How can we ensure that platforms have good risk management systems in place, without constraining innovation in scoring models?
  • How do we regulate the alignment of incentives between platforms and lenders?
  • How can we ensure a smooth resolution in case of failure?

Market power and barriers to entry

Lending-based crowdfunding platforms are entering a market dominated by large banks and face many barriers to entry. Due to high switching costs, platforms are forced to pursue a market-expansion strategy toward risky borrowers who are underserved by incumbent banks (De Roure et al. 2016, Morse 2015).

This worked during the post-crisis period, when banks needed to deleverage and cut their loan supply to creditworthy borrowers (Havrylchyk et al. 2018). But if entrants serve only borrowers who have been rejected by banks, they risk higher default rates and lose servicing fees. This means that the long-term viability of the crowdfunding business model is likely to require that banks and platforms directly compete for good borrowers.

This is complicated: incumbent banks have an informational advantage over lending-based crowdfunding platforms. Long-term relationships give banks granular data on their borrowers, that allow them to model the default risk. The scale and scope economies inherent in financial intermediation are also a significant barrier to entry. Finally, explicit and implicit government guarantees imply a funding-cost advantage. They allow large banks to provide credit at a lower interest rate than crowdfunding platforms, all other things being equal.

Regulatory challenges: Market power, and barriers to entry

  • How should we level the playing field between platforms and banks?
  • Should policymakers make it easier to switch to new technologies, and with what instruments?
  • How can we ensure that policies towards incumbent banks (such as explicit and implicit guarantees that result in lower funding costs) do not distort the level playing field between banks and crowdfunding platforms?

Institutional lenders on crowdfunding platforms

The above discussion suggests that the crowdfunding business model could potentially be more stable than banking, because it is less leveraged, less prone to runs and easier to resolve. On the other hand, most platforms allow sophisticated institutional lenders to invest because this provides an important guarantee to borrowers that their loans will be funded. But if these institutional lenders are leveraged and prone to runs and moral hazard problems, all the advantages of the lending-based crowdfunding business model could be lost.

Regulatory challenges: Institutional lenders

  • How can we ensure fair treatment of both retail and institutional investors?
  • How can we limit the interconnectedness of lending-based crowdfunding platforms and leveraged, too-big-to-fail institutions?

Concluding remarks

To sum up, lending-based crowdfunding platforms are experimenting with different business models. Therefore it is important to ensure that this experimentation happens under the eye of strong regulators and supervisors, who are able to intervene to address the potential market failures.


De Roure, C, L Pelizzon, P Tasca (2016), “How does P2P lending fit into the consumer credit market?”, Deutsche Bundesbank discussion paper 30/2016.

European Commission (2018), “Proposal for a regulation of the European parliament and of the council on European crowdfunding service providers for business”.

Havrylchyk, O and M Verdier (2018), “The financial intermediation role of the P2P lending platforms”, Comparative Economic Systems 60(1): 115-130.

Havrylchyk, O, C Mariotto, T Rahim and M Verdier (2018), “What drives the expansion of the peer-to-peer lending?”, working paper.

Havrylchyk, O (2018), “Regulatory framework for the loan-based crowdfunding platforms”, OECD working paper 1513.

Morse, A (2015), “Peer-to-Peer Crowdfunding: Information and the Potential of disruption in consumer lending”, NBER working paper 20899.

Zhang, B, T Ziegler, L Mammadove, D Johanson, M Gray and N Yerolemou (2018), 5th UK Alternative Finance Industry Report, Cambridge Centre for Alternative Finance.

Olena Havrylchyk

Automation and unemployment: Help is on the way (VOX)

Many innovations come in the shape of machines that replace workers. We hear of cars that drive themselves, of robots that perform more and more tasks, and of how artificial intelligence can replace smart jobs. These technological developments cause alarm among many, and this has intensified since the last recession that began in 2008. The recovery from the recession has been slow, and especially in creating new jobs. That is why many have called it ‘a jobless recovery.’

Interestingly, this fear of ‘technological unemployment’ is not new and has surfaced many times since the Industrial Revolution, as workers feared that new machines might drive them out of jobs. This led to the rebellion of the Luddites in 1811-1817. They were artisans who viewed in awe how the mills of the textile industry threatened their existence. They started a mutiny and the British government had to send a large army to oppress it.

The issue came up again often, especially in periods of rapid technical change. On 26 February 1928, Evan Clark wrote an article in the New York Times, titled “March of the Machine Makes Idle Hands”. He claimed that “the onward march of machines into every corner of our industrial life had driven men out of the factory and into the ranks of the unemployed”. At the time, the US rate of unemployment was only 4.2%. Around then, Keynes (1930) coined the term ‘technological unemployment’ in his famous essay “Economic Possibilities for our Grandchildren”.

The experience of the past two centuries has shown that this fear did not materialise, but the issue keeps returning to public discussion. This is partly because our memory is short and partly because some claim that ‘this time is different’ and that the risk of automation is greater. A recent publication by the World Bank discusses this claim seriously and offers ways to reduce the risk of unemployment (Chuah et al. 2018). Throughout the years, many economists were more optimistic. Keynes (1930) suggested that future technologies will increase leisure and thus will increase employment. Zeira (1998) has shown that machines that replace labour in some tasks make labour in other tasks more productive, so the demand for labour increases. In a sequence of papers, Acemoglu and Restrepo (2018a, 2018b among others) show that if technology creates new tasks in addition to automation, it increases the demand for labour and keeps the rate of unemployment from rising.

Our recent paper takes this line of research further and makes a bold contribution to the debate (Nakamura and Zeira 2018). We argue that the rate of technological unemployment will decrease in the future and will converge to zero in the end. As we show below, the mechanism we describe is different from those suggested both by Keynes (1930) and by Acemoglu and Restrepo (2018a, 2018b).

Our analysis of automation and unemployment uses a theoretical framework of production by tasks, in which technical change involves two processes. One is automation, namely shifting existing tasks from production by labour to production by newly invented machines. The second process is adding new tasks. Unlike Acemoglu and Restrepo, we do not impose any assumption on the rate of creation of new tasks, and this rate can be zero as well. As shown below, the dynamics of automation are what drives the creation of new tasks.

To explain the main result of our paper, first note that the dynamics of automation can fall into two main cases. The first is that automation does cover all tasks and stops at some finite level. In this case, the rate of automation must decline as it gets close to the upper bound, and it converges to zero over time. Since the rate of technological unemployment is proportional to the number of new automated tasks relative to the total number of labour tasks, it follows that the rate of unemployment converges to zero as well. Clearly, this is a simple case, in which our main result holds.1 The more interesting case is when automation grows without bounds, which we discuss next.

Automation grows only if producers adopt it, i.e. if they prefer capital over labour for the next tasks. We assume that newly invented machines or robots are more expensive, as we first invent the easy machines and then over time invent increasingly more complicated ones. This is a reasonable assumption that is supported by empirical studies. Producers prefer a machine for a certain task to labour only if it costs less. This means that in order to continue to adopt the ever-costlier machines, wages must rise as well. This is required to keep automation going.

In our model, the wage level depends positively on the number of labour tasks and negatively on the share of labour in output. The reason for the first positive dependence is straightforward – the larger the number of labour tasks, the smaller the number of workers who perform each task. Hence, their marginal productivity in the task is higher and this raises wages. The negative effect of the share of labour on wages is harder to explain intuitively. If wages are higher, the tasks performed by labour are more expensive and producers will buy less of them. In order to keep output at the same level, producers purchase more of the automated tasks. This increases the share of capital in output and reduces the share of labour in output. Hence, the share of labour has a negative correlation with the wage rate.

As a result, if automation is unbounded wages rise continuously, and that can happen in one of two ways: either the number of labour tasks increases rapidly (actually faster than automation) or the share of labour goes down continuously all the way to zero. If we rule out the case that the share of labour converges to zero, then the number of labour tasks must increase continuously. This means that the rate of unemployment is decreasing and is converging to zero, since it is proportional to the number of new automated tasks divided by the number of existing labour tasks. If the denominator – the number of existing labour tasks – increases, the rate of unemployment falls and converges to zero.

Hence, our study shows that the dynamics of the rate of unemployment depend crucially on the dynamics of the share of labour. Many empirical studies, the most famous of which is Kaldor (1961), have actually found that the share of labour in output is very stable, at around two-thirds, both across countries and over time. In recent decades, the share of labour has experienced a slight decline, but it is still around 60% of GDP and in any case, it is far from going down to zero. Hence, the assumption that the share of labour does not converge to zero is fully in line with the well-known stylised facts of modern economic growth. This supports our result that the rate of automation unemployment should decline over time and converge to zero.

Our paper therefore presents a strong result. Help is on the way and automation unemployment will not rise – on the contrary, it will go down to zero at the end of times. Nevertheless, we should treat this prediction with some caution. In general, economic models are better at explaining processes and mechanisms than at predicting the future. Our paper should be treated accordingly. Its main message is that although automation causes unemployment by turning labour tasks into machine tasks, it might also ignite a mechanism that reduces unemployment. Automation requires rising wages, and that requires increasing the set of labour tasks. This increase reduces the rate of unemployment. Hence, automation leads to a continuing reduction in unemployment by its own adoption mechanism. This is an important point to bear in mind when we consider the effect of automation on unemployment and on the labour market in general.


Acemoglu, D, and P Restrepo (2018a), “The Race between Man and Machine: Implications of Technology for Growth, Factor Shares and Employment”, American Economic Review 108, 1488-1542.

Acemoglu, D, and P Restrepo (2018b), “Artificial Intelligence, Automation and Work”, mimeo.

Chuah, L L, N V Loayza, and A D Schmillen (2018), “The Future of Work: Race with – not Against – the Machine”, World Bank Group, Research & Policy Briefs.

Kaldor, N (1961), “Capital Accumulation and Economic Growth”, in F A Lutz and D C Hague (eds.), The Theory of Capital, 177–222, New York: St. Martins Press.

Keynes, J M (1930), “Economic Possibilities for our Grandchildren”, in Essays in Persuasion, New York, W.W. Norton & Co., 1963, 358-373.

Nakamura, H, and J Zeira (2018), “Automation and Unemployment: Help is on the Way”, CEPR Discussion Paper no. 12974.

Zeira, J (1998), “Workers, Machines, and Economic Growth”, Quarterly Journal of Economics 113, 1091-1117.


[1] There is a sub-case of this case, in which the economy converges to an equilibrium with production by machines and robots only, with no labour. We rule out this sub-case below, as the share of labour in output goes to zero.

Hideki Nakamura, Joseph Zeira

Brexit deal debate reveals dark side to EU diplomacy (The Interpreter)

The saving grace of a nasty divorce is durable insight into the true values of the parties involved. And so, with Brexit.

The Withdrawal Agreement – which has triggered rancorous opposition in parliament and a political crisis in the UK – lays bare the diplomatic cards. Whatever its eventual fate, the way that Agreement was reached and the reaction it has now provoked provide analysts with rich insight into some of the forces now at play in Europe.

First, the anguished yelps in London show where power has prevailed. As Australia’s ex-envoy pointed out, the UK was set to become a regulatory colony of the European Union for at least 20 months. The Agreement also required UK taxpayers to foot a £39 billion (A $68 billion) divorce fee, and parliament to surrender, indefinitely, the right to legally extract itself from EU institutions thereafter.

In contriving to impose such terms, the EU revealed some unanticipated qualities: unity, discipline, and exceptional statecraft. The laurels go to the chief orchestrator, the European Commission. That furtive chink of glassware in Brussels three weeks ago now looks premature, but if the Withdrawal Agreement survives in modified terms, then November 2018 will mark the date when the Commission stepped forth as the principal power-broker in Europe.

How was this putative victory gained? Last December, the EU demanded Northern Ireland as a hostage for the trade talks: if UK and EU cannot agree on seamless trade, then a “backstop” arrangement would see Northern Ireland excised from the UK’s regulatory and customs ambit, and retained within EU jurisdiction. It is to mitigate this legal-regulatory annexation that the UK offered itself in entirety into backstop bondage.

Such diplomacy is predatory, and former foreign secretary Boris Johnson called it out. Put simply, the EU demanded that UK cede regulatory and judicial sovereignty over part of its territory as the price of a deal. In asserting this right, the Good Friday Agreement (GFA) was, proverbially, honoured more in the breach than the observance. The Withdrawal Agreement obviated a physical border in Ireland, sure, but this consequential benefit was secured by subverting the GFA’s principal provisions: acknowledgement of UK sovereignty; and constitutional change only via popular consent.

Were no qualms expressed in Brussels? Respect for sovereignty and popular consent are precisely what differentiate post- from pre-war European diplomacy, yet the Agreement respects neither and the British parliament bridled. Perhaps for the EU, the ends justified the means. If so, it’s worth examining the inevitable result of the backstop if it isn’t removed from the Agreement: UK’s forced retention in a customs arrangement and adherence to single market regulation.

To appreciate the gravity of this outcome it is necessary to grasp one fundamental point: that, although the Customs Union‒Single Market looks like a free trade area it doesn’t behave like one, at least not for UK.

Since 1998, UK’s goods exports to EU have stagnated while imports have grown at a brisk 3.2% per annum. The result: a series of chronic deficits that recur and augment like nightmares across UK trade: -£28 bn in autos; -£18 bn in food & agriculture; -£9 bn in pharmaceuticals; -£7 bn in machinery.

Scrutiny of UK’s historical trade data reveals a curious “captive market” effect. Take UK’s biggest traded sector: autos, worth £103 billion per year. Over 20 years, the EU has retained a vice-like market share of over 83% of UK imports, even as its own share of UK exports has plummeted. Meanwhile, UK’s growing dependence on EU agriculture force-feeds UK consumers on the most expensively produced food on earth.

Protected by external EU tariffs, EU producers are steadily cornering UK’s markets without commensurate reciprocity. From this, the backstop ensures no escape (without sacrificing Northern Ireland). Under the Agreement, UK’s financial services would get no special post-Brexit treatment, but even if they did, their surplus covers just one-quarter of UK’s £95 bn EU goods-trade deficit. For historians this is mercantilism; the practice of using power to capture markets. It, too, has nasty precedence in pre-war Europe.

As for the UK, enduring insights flow from reactions to the Agreement. Specifically, popular hostility to its terms proves there was always way more to Brexit than an anti-immigrant spasm. The Agreement would definitely end free movement, and yet no poll of leave voters comes close to showing a preference for the Agreement over simply walking away, with all its hazards.

The reaction amongst pro-Brexit Conservative MPs is more fascinating still. Almost all are implacably opposed. Tally up their antipathies to the Agreement – and its impediments to free trade, deregulation, cheap food, and an unfettered foreign policy – and analysts aren’t staring at swivel-eyed Tories; they are staring straight into the pupils of Gladstonian liberals who aren’t afraid of change.

So, this is a salutary moment. The rejection of the Agreement by pro-Brexit MPs reveals deep tides. To characterise Brexit, now, as populist xenophobia is as misplaced as it would be to characterise the Protestant Reformation as just an irate bout of iconoclastic statue-bashing. There’s more at stake than perceived public delinquency; specifically, a classic and compelling model of government.

As for the EU, the omens look ill. Following its rejection by Parliament, the EU has refused to renegotiate the Agreement. In its current form, the Agreement will fail. Playing for higher stakes, the EU may want it to fail. But nothing will elide the Agreement’s brutal logic or exploitative intent. Sadly, the EU’s diplomatic standards are slipping into unsavoury ways. What’s odd, though, is that none of its members object.
 Data source UK Office for National Statistics, September 2018 (November 2017 for Financial Services), with calculations made by the author, available here.

Phil Radford