by Christophe Blot

The end of 2021 was marked by growing concern among central banks about inflation[1]. As pressure on prices intensified with Russia’s invasion of Ukraine, could this change the terms of the discussion and influence future monetary policy decisions? Indeed, in February, the inflation rate reached 5.9% in the euro area and 7.9% in the US[2], well above the 2% target of the ECB and Federal Reserve. The January policy meetings suggested that a rate increase was imminent in the US and likely by the end of the year in the euro area[3]. So what is the situation today? The war between Russia and Ukraine has not only shaken up the geopolitical situation but is expected to affect the global economy, accentuating inflationary pressure, reducing household purchasing power and fuelling uncertainty. Finally, the risk of a sovereign default by Russia could also rekindle financial tensions, in particular viaa risk of contagion in the emerging countries. In this new context, one could expect greater caution and a more wait-and-see approach, as suggested in a post by Xavier Ragot. However, neither the ECB at its meeting on 10 March nor the Federal Reserve on 16 March have changed their tune. The banks remain focused on inflation.

As stated in the introductory statement of the ECB press conference on March 10, Christine Lagarde acknowledged the many uncertainties linked to the conflict’s economic repercussions. But she also stressed the strength of the economic recovery, with growth in the euro area expected to reach 3.7% in 2022 and 2.8% in 2023, according to the Eurosystem. These forecasts have been revised downwards since December 2021 by 0.5 and 0.1 points respectively. However, the ECB has decided to end its asset purchase programme (APP) more quickly, with it gradually decreasing in net terms to 10 billion euros in June. Beyond that, “the calibration of net purchases for the third quarter will be data-dependent and reflect our evolving assessment of the outlook”. In other words, net purchases should cease unless inflation and inflation expectations fall sharply[4]. 4] Recall that in December 2021, it was envisaged that purchases under the APP would continue until the third quarter of 2022. Indeed, in the short term, the shock of Russia’s invasion of Ukraine will undoubtedly translate into higher inflation, fuelled in particular by rising prices for energy and certain foodstuffs. Thus, the ECB’s inflation expectations have been revised upwards: 5.1% on average over 2022 compared to a forecast of 3.2% in December 2021. Does this mean that the ECB is planning to raise rates soon? The press release issued at its previous meeting on 3 February stated: “The Governing Council expects net purchases to end shortly before it starts raising key ECB interest rates”. Assuming that asset purchases are now scheduled to wind up in June, the likelihood of a rate hike becomes greater. A qualification is needed, however, as its 10 March press release states that, “Any adjustments to the key ECB interest rates will take place some time after the end of our net purchases under the APP and will be gradual”. So the end of purchases is definitely put forward, but now the rate hike would take place not “soon after” but “some time after”. This is still widely considered possible, although it cannot be said that it is more likely today than at the end of the 3 February meeting. Moreover, to a journalist who explicitly asked whether “some time after” ruled out the possibility of a rate hike this year, Christine Lagarde replied that no action had been ruled out and that the ECB’s communication was intended to give itself as many options as possible.

However, the ECB does seem to be focusing on inflation. Beyond the short-term inflationary shock, the ECB is looking closely at inflation one or two years hence, since this is the horizon at which a monetary policy decision affects prices. So what’s most important for the rate scenario are inflation expectations for 2023 and 2024, and not for 2022. If long-term inflation converges to or exceeds the 2% target, the ECB will surely raise rates as the need for monetary support fades[6]. According to the latest forecasts, the ECB expects inflation to reach 2.1% in 2023 and 1.9% in 2024, which are close to the target (Figure 1).

With inflation close to target, growth robust and unemployment falling, the prospect that monetary policy will be normalized may seem fitting. However, note that higher inflation is being driven largely by food and energy prices. Apart from these two components, the ECB expects inflation to be 1.8% in 2023 and 1.9% in 2024[7]. Under these conditions, the ECB is in a dilemma, with a shock that is resulting in higher inflation but also slower growth, which could delay the return of growth to its potential[8]. If inflation remains essentially driven by energy and food prices, then a rate hike would not be effective in reducing it but would accentuate the negative shock on the economy. So while the ECB’s primary objective remains inflation, tightening monetary policy is worthwhile only if it helps to achieve this objective. In the current context, the ECB will have to find the right mix between on the one hand fighting against a risk of runaway inflation that is linked to possible second-round effects and on the other risking undermining the recovery.

From this point of view, the situation of the US is different even if, as in the euro area, the FOMC members have revised the US growth forecast for 2023 downwards and the inflation forecast upwards. The US economy is probably less exposed to the shock of the war. The main difference with the euro area, however, is the level and nature of the inflation. Indeed, the change in inflation is not only a consequence of pressure on energy prices, as the year-on-year increase in the underlying consumer price index was 6.4% in February, compared to 2.7% in the euro area. Moreover, wages also seem to be taking off, reflecting tensions in the US labour market and thus a much higher risk of overheating than in the euro area, which would justify faster and probably stronger action by the Federal Reserve[9]. It is therefore not surprising that the FOMC members were broadly in favour of a quarter-point increase in the federal funds rate at the meeting held on 16 March. This hike in the monetary policy rate had been announced implicitly at the previous meeting and was widely anticipated. This trend could even pick up pace since, at the end of the FOMC meeting scheduled for 15 June, according to the FED watchers, there is a 55% probability that the rate will reach 1.25% and a 33% probability that it could hit 1.5% (Figure 2)[11]. However, even if higher rates seem more justified in the United States, the Fed will also have to take into account the impact of interest rates on medium-term debt dynamics. Given the level of public debt (130% in 2021 versus 109% in 2019), close coordination of monetary and fiscal policies is likely necessary to reconcile the objectives of fighting inflation, maintaining growth and gradually reducing public debt. As Gilles Dufrénot reminds us, debt reduction after the Second World War was accompanied by a low real interest rate strategy[12].

---

[1] See the OFCE post of 20 January 2022.

[2] The consumption deflator, an indicator monitored by the Federal Reserve, was rising by 6.1% year-on-year in January 2022.

[3] Note that in the UK, January inflation was 5.5% and the Bank of England had already raised its key interest rate twice.

[4] The flow of asset purchases by the ECB under the APP leads to increasing the size of its balance sheet. Terminating the programme does not imply a cessation of purchases but rather the end of increases in the size of its balance sheet. Thus, the ECB will replace maturing assets with purchases that stabilize the balance sheet.

[5] In December 2021, the ECB envisaged net purchases of 30 billion euros in the third quarter of 2022.

[6] It is indeed conceivable that, given the current level of interest rates, a small hike would not contribute to slowing down activity but would reflect less support.

[7] Recall that since July 2021 the ECB has communicated a new inflation target of 2%, as opposed to its previous “close to but below 2%”. However, the measure of inflation remains the HICP, an indicator that includes energy and food prices. See Blot, Bozou and Hubert (2021) for more detail [in French].

[8] Indeed, central banks generally react to the gap between actual inflation and the target and the gap between the level of activity and potential GDP. Thus, rapid growth does not indicate that activity is exceeding its potential. Indeed, according to the OECD, this growth gap should still be negative in 2023 (-0.3%). However, this estimate does not take into account the impact of the economic shock linked to the war in Ukraine.

[9] See Domash and Summers (2022) for a more in-depth analysis of the tensions in the US labour market. Although the unemployment rate has not yet returned to its early 2020 level, other indicators such as the employee resignation rate and the job vacancy rate point to greater pressure.

[10] All but one member voted in favour of this increase, with the dissenting voice in favour of a half-point increase.

[11] A meeting is also scheduled for 4 May, at which there is a 58% probability of a rate hike of 0.25 points and a 42% probability of 0.5 points.

[12] See Reinhart and Sbrancia (2015) for a more detailed analysis of public debt reduction after 1945 in the industrialized countries.

by Gregory Verdugo

An unprecedented crisis

Since the war in Ukraine started, unprecedented numbers of refugees have poured across the country’s borders. As of 15 March, the UN High Commissioner for Refugees counted more than three million refugees who have crossed the border since the Russian offensive began on 24 February. In just three weeks, the number of refugees has surpassed the year-long peaks reached during the 2015 and 2016 migration crisis. The number already exceeds the total number of refugees that followed the Bosnia and Herzegovina war in the years 1993-1995.

A specific status

In order to avoid having asylum applications filed in more than one European country, the Dublin III Regulation (2013) requires refugees to apply for asylum only in the first country through which they entered the EU. This regulation aims to encourage border states to better monitor their borders but also both to clarify which country is responsible for examining the asylum application and to prevent attempts at “asylum shopping”.  During the 2015 migration crisis, this system disproportionately placed the burden of receiving refugees on countries with a Mediterranean border that were directly on the route of Syrian and Afghan refugees. Countries such as Greece and Malta were soon overwhelmed by asylum applications, which far exceeded their processing capacities.

Faced with the massive and rapid flow of Ukrainian refugees, the European Union, aware of the inadequacy of the Dublin Regulation, took unprecedented steps. The Temporary Protection Directive, drawn up in 2001, was activated for the first time on 4 March 2022, following its unanimous adoption by the EU Council of Interior Ministers on a proposal from the EU Commission.

Temporary protection offers Ukrainian refugees a right of residence for one year, which can be extended up to three years. Beyond the right of residence, temporary protection also provides access to education, which is crucial given the number of displaced families, and it guarantees access to social and medical assistance and the right to family reunification.

Temporary protection simplifies the reception of Ukrainian refugees by preventing the clogging of asylum systems. In some countries, processing asylum applications can take several years before a final decision is reached. Even if a fast-track procedure had been introduced, it would have been difficult to avoid overloading the asylum system and lengthening the processing time for the large number of Ukrainian refugees. Such long delays penalise the refugees. Uncertainty about the possibility of staying in the host country reduces in particular the incentive to establish links with the country and to learn the language (Hainmueller et al., 2016).

Another advantage of temporary protection is that it allows Ukrainian refugees immediate access to the EU labour market. Only four countries in the EU have previously allowed asylum seekers immediate access to the labour market. All the rest restrict access to employment for periods of between two and 12 months, and sometimes indefinitely. Recent studies have shown that work bans are particularly costly, not only because people do not contribute immediately to the economy, but also because the bans have a persistent negative effect on asylum seekers’ subsequent employment after they have finally been granted refugee status (Fasani et al., 2021).

The challenge of economic integration

The scale of future arrivals and the length of stay of refugees will depend both on what happens with the conflict and on Ukraine’s post-conflict economic prospects. While not all refugees will want to stay in the EU, the large-scale destruction already witnessed suggests that the country’s post-conflict economic difficulties may prompt many refugees to prolong their stay or even to settle down. The return of refugees could also be jeopardised by the lack of security in certain areas or in the country as a whole. It is therefore likely that a proportion of refugees will stay in the EU for a long time, if not permanently, as happened with Yugoslav refugees long after the conflict ended (Bahar et al., 2022).

The economic integration of refugees poses specific challenges (Verdugo, 2019). Most studies suggest that, at least initially, refugees encounter more difficulties than economic immigrants in finding employment and integrating into the host country’s labour market (Dustmann et al., 2017). This is because economic migrants prepare for migration, so those who migrate are positively selected, i.e. it is those who are best prepared and most capable of succeeding within their home population who try their luck abroad. Economic migrants are also more likely than refugees to master the language of the host country and to benefit from solidarity networks that help them to integrate economically (Borjas, 1987). In contrast, the migration of refugees is not economically motivated. They are forced to migrate in order to escape physical insecurity, and they do so in an emergency. Refugees are more often workers whose knowledge is less valuable in their host country (Chiswick, Lee and Miller, 2005).

On the other hand, refugees, unlike economic migrants, face uncertainty about the possibility of returning to their country of origin. Their migration is more likely to take place over a longer time horizon than that of economic migrants, which may encourage them to build long-term relationships with the host country. Cortes (2004) finds that, in the United States, while refugees initially face greater economic hardships, they tend to catch up with economic migrants in the longer term.

How to allocate refugees

The burden of hosting refugees has always been unevenly distributed (Huertas Moraga and Hagen, 2021). The Ukrainian crisis is no exception. Most Ukrainian refugees are currently in the countries bordering Ukraine and, as of 15 March, over 60% were in Poland. As in the 2015 migration crisis, EU countries face the challenge of spreading their reception over a number of countries so that the cost doesn’t fall on a small number of countries and exhaust their good will.

Despite the adoption in 2015 of a European Agenda for Migration, which highlights the benefits of cooperation, progress in this respect has been limited. The Pact on Migration and Asylum proposed by the European Commission in September 2020 has yet to be discussed by the EU Parliament and Council. In this draft, the European Commission proposes to introduce mandatory quotas based on GDP and population size, with some flexibility. A new proposal is that each country can choose either to take in refugees or to contribute to the cost of returning migrants whose asylum applications have been rejected.

Other innovative proposals are also circulating. In an influential article, Fernández-Huertas Moraga and Rapoport (2014) propose the introduction of quotas that are then tradable on a quota market between EU countries. So if a country wants to reduce its quota and take in fewer asylum seekers, it can pay another country to receive more. Fernández-Huertas Moraga and Rapoport (2014) also propose the introduction of a system that could match asylum seekers’ preferences with each state’s preferences. Asylum seekers would thus declare which countries they prefer, and the countries would identify their preferences for different categories of asylum seekers. A centralised allocation algorithm would allocate asylum seekers based on the two groups’ respective preferences. While governments have always been reluctant to offer asylum seekers greater choice, the EU Commission is nevertheless proposing to take into account their preferences and try to receive them in countries where they have “meaningful links”.

Whatever system is put in place to distribute the Ukrainian refugees, they are currently free under temporary protection to move between European countries and thus to choose their preferred destination. If reception quotas are introduced, it is not clear how effective they will be unless refugees’ mobility is restricted. However, it seems difficult to move refugees in an authoritarian manner to countries that they have not chosen and where they have no ties and may have difficulty integrating. In the short term, the most credible solution seems to be to combine compensation for the countries that receive the most refugees with incentives to settle in countries that do not receive so many.

par Gregory Verdugo

Une crise inédite

Depuis le début de la guerre en Ukraine, un nombre inédit de réfugiés ont afflué aux frontières du pays. Le Haut-Commissariat des Nations unies pour les réfugiés recensait au 15 mars plus de 3 millions de réfugiés ayant franchi la frontière depuis le début de l’offensive russe le 24 février. En à peine trois semaines, le nombre de réfugiés surpasse les pics atteints sur une année entière au moment de la crise migratoire de 2015 et 2016. Il surpasse déjà le total de réfugiés qui ont suivi la guerre de Bosnie-Herzégovine dans les années 1993-1995.

Un statut spécifique

Afin d’empêcher de déposer une demande d’asile dans plusieurs pays européens, le règlement de Dublin III (2013) impose aux réfugiés de demander l’asile uniquement dans le premier pays par lequel ils sont entrés dans l’UE. Ce règlement vise à inciter les états frontaliers à mieux surveiller leurs frontières mais aussi à clarifier le pays responsable de l’examen de la demande d’asile et empêcher les tentatives d’« Asylum shopping ».  Lors de la crise migratoire de 2015, ce système avait fait peser de manière disproportionnée le poids de l’accueil des réfugiés sur les pays ayant une frontière méditerranéenne et qui se trouvaient sur la route des réfugiés syriens et afghans. Des pays comme la Grèce ou Malte ont vite été dépassés par le traitement des demandes d’asiles trop nombreuses pour leurs capacités.

Face au flux massif et rapide de réfugiés ukrainiens, l’Union européenne, consciente du caractère inadapté du règlement de Dublin, a réagi de manière inédite. La directive de protection temporaire, élaborée en 2001, a été pour la première fois activée le 4 mars 2022 à la suite de son adoption à l’unanimité par le Conseil des ministres de l’intérieur de l’UE sur proposition de la Commission.

La protection temporaire offre aux réfugiés ukrainiens un droit de résidence d’un an qui peut être prolongé jusqu’à trois ans. Au-delà du droit de séjour, la protection temporaire offre l’accès à l’éducation, qui est crucial étant donné le nombre de familles déplacées, et elle garantit l’accès à l’aide sociale et médicale et le droit au regroupement familial.

La protection temporaire simplifie l’accueil des réfugiés ukrainiens en évitant l’engorgement des systèmes de demande d’asile. Dans certains pays, le traitement des demandes d’asile pouvait durer plusieurs années avant qu’une réponse définitive soit obtenue. Même en cas de mise en place d’une procédure accélérée, il aurait été difficile d’éviter l’engorgement des services de demande d’asile et un allongement des délais de traitement face au nombre de réfugiés ukrainien. Or ces longs délais pénalisent les réfugiés. L’incertitude sur la possibilité de séjourner sur le territoire diminue, notamment les incitations à nouer des liens avec le pays d’accueil ou apprendre la langue (Hainmueller et al. 2016).

La protection temporaire a pour avantage d’autoriser immédiatement les réfugiés ukrainiens à accéder au marché du travail de l’Union. Dans l’Union européenne, seuls quatre pays autorisaient les demandeurs d’asile à accéder immédiatement au marché du travail. Les autres pays restreignaient l’accès à l’emploi sur des périodes comprises entre 2 à 12 mois et parfois même indéfinies. Des travaux récents ont montré que les interdictions de travailler sont particulièrement coûteuses, non seulement parce qu’elles ne contribuent pas immédiatement à l’économie, mais aussi parce qu’elles se traduisent par des effets négatifs persistants sur l’emploi ultérieur des demandeurs d’asile lorsque leur statut de réfugié leur est finalement accordé (Fasani et al. 2021).

Le défi de l’intégration économique

L’ampleur des futures arrivées et la durée du séjour des réfugiés dépendent de l’évolution du conflit et des perspectives économiques de l’Ukraine à l’issue du conflit. Même si tous les réfugiés ne voudront pas rester dans l’Union européenne, l’étendue des destructions déjà constatées suggère que les difficultés économiques du pays après le conflit pourrait inciter de nombreux réfugiés à prolonger leur séjour ou même s’installer. Le retour des réfugiés pourrait également être compromis par l’absence de sécurité dans certaines régions ou le pays entier. Il est donc vraisemblable qu’une partie des réfugiés séjourne de manière prolongée si ce n’est permanente dans l’Union européenne comme cela a été observé pour les réfugiés yougoslaves bien après la fin du conflit (Bahar et al., 2022).

L’intégration économique des réfugiés pose des défis spécifiques (Verdugo, 2019). La plupart des études suggèrent que les réfugiés ont, au moins initialement, plus de difficultés que les immigrés économiques à être employés et intégrés dans le marché du travail de leur pays d’accueil (Dustmann et al. 2017). En effet, les migrants économiques préparent leur migration et ceux qui migrent sont sélectionnés positivement, c’est-à-dire que ce sont les mieux préparés et les plus capables de réussir au sein de leur population d’origine qui tentent leur chance à l’étranger. Plus souvent que les réfugiés, les migrants économiques maîtrisent la langue du pays d’accueil et bénéficient de réseaux de solidarités qui les aide à s’intégrer économiquement (Borjas, 1987). Au contraire, la migration des réfugiés ne répond pas à des motifs économiques. Elle est subie afin d’échapper à l’insécurité physique et s’effectue dans l’urgence. Les réfugiés sont plus souvent des travailleurs dont les connaissances sont moins valorisables dans leur pays d’accueil (Chiswick, Lee et Miller, 2005).

D’un autre côté, contrairement aux migrants économiques, la possibilité de migration retour dans le pays d’origine des réfugiés est incertaine. Leur migration se place plus souvent sur un horizon plus long que celui des immigrés économiques, ce qui peut les inciter à nouer des relations bâties sur le long terme avec le pays hôte. Cortes (2004) constate ainsi qu’aux États-Unis, si les réfugiés rencontrent initialement plus de difficultés économiques, ils tendent à rattraper les migrants économiques à plus long terme.

Comment répartir les réfugiés

La charge de l’accueil des réfugiés a toujours été répartie de manière inégale (Huertas Moraga et Hagen, 2021). La crise ukrainienne ne fait pas exception. La plupart des réfugiés ukrainiens se trouvent actuellement dans les pays frontaliers de l’Ukraine et, à la date du 15 mars, plus de 60% se trouvent actuellement en Pologne. Comme lors de la crise migratoire de 2015, les pays de l’UE font face au défi de répartir leur accueil sur plusieurs pays afin d’éviter que le coût ne repose sur un petit nombre de pays dont la bonne volonté risque de s’épuiser.

Malgré l’adoption en 2015 d’un Agenda européen pour les migrations qui souligne les bénéfices de la coopération, les progrès ont été limités. Le pacte sur la migration et l’asile proposé par la Commission européenne en septembre 2020 doit encore être examiné par le Parlement et le Conseil de l’Union. Dans ce projet, la Commission européenne propose d’instaurer des quotas obligatoires basés sur le PIB et la taille de la population tout en introduisant de la flexibilité. Une proposition nouvelle est que chaque pays peut soit choisir d’accueillir des réfugiés, soit participer au coût du retour au pays des migrants dont la demande d’asile a été rejetée.

D’autres propositions innovantes circulent. Dans un article influent, Fernández-Huertas Moraga et Rapoport (2014) proposent également d’instaurer des quotas mais de les rendre échangeables au travers d’un marché de quotas entre pays de l’Union. Ainsi, si un pays veut réduire son quota et accueillir moins de demandeurs d’asile, il doit payer un autre pays pour qu’il en accueille davantage. Fernández-Huertas Moraga et Rapoport (2014) proposent également d’instaurer un système d’appariement afin de faire correspondre le désir des demandeurs d’asile aux préférences de chaque État. Dans ce système, les demandeurs d’asile expriment leurs préférences pour les pays, et les pays leurs préférences pour différentes catégories de demandeurs d’asile. Un algorithme d’allocation centralisé se chargerait d’allouer les demandeurs d’asile en prenant en compte leurs préférences et celles des pays. Si les gouvernements ont toujours été réticents à offrir plus de choix aux demandeurs d’asile, la Commission propose néanmoins de prendre en compte leur préférences et d’essayer de les accueillir dans les pays où ils ont des « meaningful links ».

Quel que soit le système qui sera mis en place pour les répartir, les réfugiés ukrainiens sous protection temporaire sont pour l’instant libres de se déplacer entre pays européens et ainsi de choisir leur destination préférée. Si des quotas d’accueil sont instaurés, leur effectivité est incertaine à moins de restreindre la mobilité des réfugiés. Or il apparaît difficile de déplacer les réfugiés de manière autoritaire dans des pays qu’ils n’ont pas choisi et où ils n’ont pas de liens et risquent d’avoir du mal à s’intégrer. À court terme, le plus crédible semble ainsi de combiner des dédommagements pour les pays qui accueillent le plus de réfugiés à des incitations à l’installation dans les pays n’en accueillant pas beaucoup.