In theory, at least three different migration flows can be distinguished between Russia and the new abroad:

• Forced migration strictu sensu from areas of ethnic and civil wars (i.e. Abkhazia in Georgia, Chechnya in Russia, Tajikistan) or resulting from grassroots inter-ethnic riots and pogroms (i.e. in 1989–1990 in Uzbekistan against Meshketian Turks)
• Post-empire repatriation of ethnic Russians to Russia, as well as of other nationalities from Russia to their homeland.
• Voluntary(economic) migrations.

Yet, as anticipated, in practice it is extremely difficult to determine the boundaries between these categories, particularly for ethnic Russians. In general, analytical distinctions are problematic, for different types of factors are entwined in determining mobility decisions, but this is even more the case in the post-Soviet context, given the coincidence of socioeconomic crisis, ethnic unrest and the break-up of a multinational state.

One could argue, as it has been often done by Russian politicians, that Russians are not voluntarily repatriating to their homeland but are forced to do so by nationalising state policies reflected in discriminatory language and citizenship laws. On the other hand, it is also a question of discussion to what extent the return of Russians from the economically depressed areas of Central Asia is ethnically and not economically driven. Russian experts interviewed by the author in Moscow claim that, even when Russian migrants are coming from areas of severe economic crisis, they should still be seen not as economic migrants but as forced migrants because of the allegedly different ways in which such conditions affect them and the members of the titular nationalities due to nationalising policies.

The first inflows into Russia of individuals escaping ethnic conflict,¹³ mainly into the southern provinces of Krasnodar and Stavropol, date back to the late 1980s after the outbreak of the Nagorno–Karabakh conflict and the Ferghana valley pogroms against Meshketian Turks (Ter-Sarkisyants 1994; Yamskov 1995). Since then, a large number of individuals have been forced to migrate (mostly to Russia) as the result of a series of ethnic conflicts and a civil war such as.¹⁴

• The Nagorno–Karabagh war between Armenians and Azeris (started in 1988, armed hostility stopped in 1994, but no agreement has yet been reached);
• The inter-ethnic clashes between Azeris and Armenians in towns of Armenia and Azerbaijan, caused by the above mentioned war (1989–1990);
• Pogroms by Uzbeks against Meshketian Turks in Ferghana valley (1990);
• Inter-ethnic clashes between Kyrgyz and Uzbeks in the Osh region (1990);
• Anti-Armenians and Anti-Russian demonstrations and attacks in the Tajik city of Dushanbe (1990);
• Separatist wars in Georgia, between Georgians and South Ossetians (1989–1992) and between Georgians and Abkhazs (1992–1994, armed hostilities have stopped, but no agreement has yet been reached);
• Transdniestr war in Moldova between Russian-speaking separatists and Moldovans (Spring–Summer 1992);
• Conflict between North Ossetians and Ingush for the Prigorodnyi district in the Russian North Caucasian region (November 1992);
• Chechen war (1994–1996).

Since 1992, the FMS has kept records of the number of people granted the official status of ‘forced migrant’ or ‘refugee’ (the juridical distinction between the two will be discussed in more detail later, see 2.1; in brief, ‘forced migrant’ status is reserved for those entitled to Russian citizenship and ‘refugee’ status to all others). The data registered by the FMS are reported in Table 5.

According to these data, there were 1,147,076 people in Russia officially registered as 'forced migrants' or 'refugees' at the end of 1996. This figure is not, however, an accurate indicator of the total number of strictly defined forced migrants who have arrived in Russia since 1988. First, it does not include those who arrived before the FMS started to operate in the summer of 1992. Secondly, as officials of the FMS declared to the author, it is estimated that only between 20–30 percent of those potentially entitled to the status actually apply for it. Finally, many of those registered could probably be better described as repatriates: those coming from Abkhazia where a violent war lasted for several months were clearly forced out by the event, but it is much more dubious that all those arriving from republics where neither open ethnic war nor grassroots inter-ethnic clashes occurred qualify for forced migrant status.

There are numerous other more or less reliable estimates of the total number of more strictly defined forced migrants present in Russia. According to UNHCR estimates reported by Slater (1994: 41), by 1993 about 2 million people had arrived in Russia from the Transcaucasian republics of Armenia, Georgia and Azerbaijan alone. On the other hand, Goskomstat estimates that 20 percent of all immigrants arriving from the new abroad are forced migrants (Komarova and Tishkov 1996: 36), a ratio which, if applied to the 1989–1996 period, would put their total number at only about 1.4 million.¹⁵

In light of such inaccuracies in the quantitative assessment of this phenomenon, the data in Table 5, despite its limitations, gives a qualitative indication of the flow. It shows how in the early years the most numerous arrivals were from Georgia, Azerbaijan, and Tajikistan, whereas after the beginning of the war in Chechnya internal refugees from the Russian Caucasus almost reached the level of those from Tajikistan. With respect to the war in Chechnya, FMS head Tatyana Regent has estimated that about 480,000 individuals fled the republic in 1995 whereas 117,000 had already left between 1992 and the start of the war.¹⁶

Let us start analysis of this flow, by far the most significant both quantitatively and for its political and symbolic implications, by looking at the overview offered by Table 6.

Between 1989 and 1996, about 3.7 million Russians arrived in Russia from the new abroad, while 1.5 million went in the opposite direction, resulting in a net balance of a little more than 2.4 million. It should be noted that Russians, though the most numerous, were only a part of the total net migration from the new abroad (3.4 million); about 1 million were immigrants of other former Soviet nationalities.

The movement of Russian settlers from the centre of Muscovite Rus' toward the huge territory which was later to constitute the Tsarist Empire and then partially the Soviet Union started in the XVI century (Arutyunyan 1992: 13). While in the second half of the seventeenth century 90 percent of Russians still resided in the four districts historically constituting Muscovite Russia – Central-industrial, Central-rural, Northern, and of the lakes – by 1917 this percentage had dropped to less than 50 percent (Bruk and Kabuzan 1982: 15).

This centrifugal movement of Russians continued and increased during the Soviet period (Arutyunyan 1992: Chapter 1), but as early as the 1960s signs of a partial reversal of the flow from the centre to the periphery became manifest. As shown in Table 7, the percentage of Russians as compared to the total increased steadily from 1926 to 1989 in Belarus, Moldova and Ukraine and from 1959 to 1989 in the Baltic republics, but started to decrease in the 1960s and 1970s in the Transcaucasian and Central Asian republics. Naturally this decrease is partially due to the higher birth rate of Transcaucasian and Central Asian titular nationalities as compared to that of Russians residing in those republics. On the other hand, if we combine the data of Table 7 with those on Soviet inter-republican net migration reported in Table 8, we can conclude that the decrease in the percentage of Russians out of the total population of Transcaucasian and Central Asian republics has also been determined by the outmigration of Russians from these republics – an outmigration which, according to several authors (Arutyunyan 1992: 17–51; Dunlop 1993; Kaiser 1994: passim; Komarova and Tishkov 1996; Perevedentsev 1993; Subbotina 1992, 1995, 1996; Topilin 1992), is explained by a combination of demographic, socioeconomic and ethnic factors.

In Table 8, the 15 Soviet republics (Russia included) have been divided into four groups according to the dynamics of their inter-republican net migration. The republics of the first group, including the Baltic republics and Ukraine, have registered net migration inflows throughout the period considered and continued to receive the immigration of Russians until the late 1980s.

For Azerbaijan and Georgia in the second group, the outmigration of Russians has been going on since the beginning of the 1960s, while for the Central Asian republics included in the third group the outmigration started in the 1970s (Kazakhstan, Kyrgyzstan and Turkmenistan) or the 1980s (Tajikistan and Uzbekistan). For the other two republics in the third group, Armenia and Moldova, the sign of net migration has a different meaning. Since both nationalities, but particularly Armenians, are known to be among the most mobile of all Soviet nationalities, behind the Russians, the negative net migration of the two republics is mainly due to their outmigration from the homeland rather than to the outmigration of Russians. In the fourth group we find Russia whose inter-republican net migration trend shows quite clearly the partial reversal of the centre-periphery flow started in the 1970s (the same, though to a smaller degree applies to Belarus).

After this brief historical excursus, Russian repatriation can now be considered more analytically, namely by looking at how the size of the inflow varies, in both absolute and relative terms, depending on the republic of the new abroad from which it originates.

Table 9 gives a synthetic overview of the distribution of the Russian diaspora across the former Soviet republics in 1989. Ukraine and Kazakhstan were the two republics with by far the highest number of Russian residents (respectively 44.9 percent and 24.6 percent of the total Russian diaspora). Yet Russians also represented a relevant quota of the total population in Estonia, Latvia and Kyrgyzstan.

The absolute figures for the 1990–1996 period tell us that by far the highest number of Russians returned from Kazakhstan, followed by Uzbekistan and Ukraine. These data, however, must also be read in relative terms, that is in terms of the initial number of Russian in each republic in 1989, as is done in Table 11, where the total net outmigration over the period 1990–1996 is divided by the total Russian population residing in each republic in 1989.

If we go back to the overview of migration flows between Russia and the new abroad (Table 6), it can be seen that Russia received a net inflow of about 1 million immigrants of other former Soviet nationalities between 1989 and 1996. Tables 12 and 13 give an interesting picture of the movements of non-Russian titular nationalities between Russia and the new abroad.

Most nationalities, with the exception of Armenians, registered net outflows out of Russia between 1990 and 1992, but net inflows between 1993 and 1996 (Table 12). In addition, the data on total immigration into Russia for 1993–1996 reported in Table 13 show that that immigrants of titular nationality arrive in Russia both from their own and from other republics. When leaving their homeland, it is reasonable to assume that non-Russian immigrants to Russia are not driven by ethnic factors but rather by socioeconomic incentives (with the exceptions of Georgia and Tajikistan where the ethnic and civil wars have also forced the titular nationalities to flee to Russia).

The same line of reasoning can, in my view, be applied to the outmigration of Russians from Ukraine and Belarus which started in 1993. Because it is unlikely that strong 'ethnic' push factors are at work in the two Slavic republics, Russian outmigration from Ukraine and Belarus is probably motivated to a large extent by socioeconomic factors. These preliminary considerations seem to confirm that to some extent migrations from the new abroad into Russia are purely socioeconomic