Bourdieu describes reconversion strategy as "the set of outwardly very different practices whereby individuals or families tend, unconsciously or consciously, to maintain or increase their assets and consequently to maintain or improve their position in the class structure... [and] an aspect of the permanent actions and reactions whereby each group strives to maintain or change its position in the social structure, or, more precisely - at a stage in the evolution of class societies in which one can conserve only by changing to change so as to conserve." (Bourdieu 1986)

This sophisticated, dynamic and finely-tuned Marxist model of class struggle has but one quite serious shortcoming - it is difficult to test empirically. Nonetheless, Bourdieu's congenial approach - to conserve by changing - was too attractive not to try to apply it when I developed my model to compare the pre- and postmigration social positions of Transylvanian forced migrants in Hungary.¹

In comparison to Bourdieu's approach, my model is greatly simplified. In my model: - wealth and job status (collectively: 'well-being') are maintained by reconversion, and not (the more complex) social position;
- only short-term pre-migration and post-migration periods are analysed, as opposed to long-term historical trends; and finally,
- the movement of individuals between two societies (that are assumed to be structurally unchanged in the short run) elicits reconversion strategies and not structural changes in social institutions.

In effect, our analysis is a sort of mobility research. In this approach, emigration marks a watershed in the subject's life, splitting it into two distinct phases.²

This watershed represents a break in the possession of material goods because forced migrants can only transfer movable and hideable property from their pre-migration to their post-migration lives.

Money, for example, can be smuggled across borders in the course of migration. It cannot, however, be smuggled in large amounts without a great deal of risk. In addition, even if it is successfully transferred, it is likely to lose its value. Gold and other Ireasures such as rare stamps, jewels, etc. provide a relatively easy way to transfer wealth. Albeit their compactness, such items were rare among our subjects. A foreign bank account also may provide an excellent means to transfer wealth in the course of migration, but foreign bank accounts were also extremely rarely held by our sample. Durable goods are, by definition, very inappropriate items to transfer, with the exception of cars - in which case the item being transferred simultaneously provides the conduit for transfer. To sum up, our model assumes that there is a severe limit to the maintenance of pre-migration wealth in the course of migration.³

If this assumption is true then we have identified a situation whereby it is impossible to maintain or improve one's material well-being by directly transferring material assets from the pre- to the post-migration phase. And if, through empirical research, we also find that wealth is preserved or even increased in the course of migration, then such an increase must have occurred through the conversion of pre-migration wealth into non-material assets, which could subsequently be reconverted in the post-migration period because of the structural similarities between the economic institutions in the two countries.

If we accept what has been said above, we need to chart the ways people convert their wealth. There are several potential reconversion strategies. In the model I identified three forms: cultural capital, network capital and direct investment into migration.

Cultural capital (education, knowledge of foreign languages, etc.) and network capital (hospitality and generosity, friendships and well-maintained relationships with relatives) - being non-tangible and invisible - are very appropriate assets to transfer. There is no danger that border guards will take them away. A practically infinite quantity of these assets can be stored and, last but not least, there is a good chance that they will not lose their value in the course of migration.

Direct investment into migration, represented by a range of strategies from bribing border guards to purchasing diplomas or visas, can increase the market value of human capital. Direct investment can reduce migration risks - risks which can be lethal.

In the analysis that follows we address the following questions:

- Were forced migrants from Transylvania able to maintain their standards of well-being in the course of migration?
- Did capital reconversion really play a significant role in this process?
- What forms of capital reconversion were the most important in maintaining wealth in migration?⁴

Besides the characteristics mentioned above, there are a few other sociologically relevant features of forced migration:

1. The degree of planning: Emigration may be unplanned at limes of natural disasters or war. However, emigration can also be carefully and strategically prepared for. The following quotations provide examples of such planning:

"Of the seven of us, Dad chose me to go and help the rest" (Forró-Havas 1988:90).

"We dreamed of leaving for weeks. I kept thinking that this little child would have a better future in Hungary than in Romania" (Forró-Havas 1988:6).

"We thought for a long time about leaving everything behind - the house, our parents to try and make it in Hungary. We finally decided to go. My mother didn't even know about our plans. But she came over a few minutes before we were about to leave, as if she suspected something. When I saw her coming to our house I hid all our bags under the bed" (Forró-Havas 1988:12).

As can be seen, forced migration can be part of a strategy extending over several generations. It is not rare for parents to take the risks inherent to emigration in their children's interest.

2. The extent of risk: In socioeconomic terms, forced migration is a risky venture. Sociologically it represents a crisis - a crisis which is assumed to be temporary. Forced migration represents a crisis because it presents the migrant with the unknown.

Forced migration is a challenge for which one cannot adequately prepare - a situation which will (no matter how well one prepares) inevitably be accompanied by unanticipated consequences.

This is primarily due to the fact that the forced migrant is unfamiliar with the new situation. In the new milieu he/she has no well-established networks, little access to information, no control over the material conditions, and (perhaps most shocking the migrant) the experiences accumulated in the 'previous life' are no longer valid. To top it off, this disorientation happens at a time when such networks and information are particularly needed because of the fluidity and uncertainty of the situation.

In other words, forced migration presents the migrant with a situation in which there are fewer resources and more stress than usual:

3. The depth of the crisis is assumed to be basically different for migrants crossing the border with valid passports (hereafter, to be referred to as 'legal forced migrants') than for those crossing illegally for the following reasons:

- Legal forced migrants can spend some time (until their tourist visa expires) collecting information about life in Hungary. They can test the viability and usefulness of personal networks and discover which agencies to apply to for formal social support. They may even be able to find housing and employment some time before their final move. During their lawful stay, potential forced migrants can do much to minimise the depth and duration of the crisis to come.

- Forced migrants arriving through legal channels can always decide to return to their original home without suffering negative consequences. In this sense, their lawful stay allows them to put off making their decision final. They can test their possibilities several limes before finally settling in the host country.

- Legal forced migrants are able to rescue more of their pre-migration assets (especially if they cross the border by car) than their clandestine counterparts.

4. Finally, forced migration by definition is different from mere migration in two respects: First, the risk of persecution is a fundamental reason for leaving the home country. Second, the forced migrant has no plans to return to his home. Consequently, compared to migration, forced migration is more characterised by a push to emigrate than a pull to immigrate, and is therefore likely to be unidirectional.

The four sociological features of forced migration noted above are important because: - we can only use the reconversion paradigm if the migrant has had some time to prepare migration, since some elements of capita conversion take time to carry out (bribing of officials, organisation, paper work, etc.);

- we can only consider the mere matching of pre- and post-migration standards of living to represent a successful reconversion of material capital if the post-migration situation is assumed to be a crisis.
 

The data

Our analysis is based on data collected in the spring of 1989. The selected sample of 1367 was representative of the roughly 15,000 forced migrants living in Hungary at that time - the first wave of Transylvanian migration.

The majority of the forced migrants had come to Hungary 1-6 months prior to our data collection. Further, our analysis covered only ethnic Hungarians between 19-35 years of age (N = 998). We limited our sample in this way to increase its cultural and life-cycle homogeneity.

In what follows, as I have described Transylvanian forced migrants in detail elsewhere (Sik 1992), I will only analyse the features needed to identify the sociological dimensions of forced migration.

1. 84% of the forced migrants had made the decision to migrate in Romania (as opposed to deciding to stay in Hungary in the course of a visit here), and 48% had carefully prepared for migration (withdrew their money from the bank, paid off their debts, found homes for their children, etc.).

2. We can provide only indirect evidence that migration presented a crisis for the forced migrants, and that this crisis largely arose because of the unfamiliarity of the new environment. For example:

- only 6% of the forced migrants had been able to purchase homes, the overwhelming majority were subtenants or lived in workers hostels;
- 40% had experienced hostility from native Hungarians;
- although all of them had jobs, and their incomes were comparable to Hungarians' incomes, only 23% were involved in the second economy.

3. Table 4 shows that both the pre- and post-migration conditions of legal forced migrants were better than those of illegal forced migrants.

4. As regards our assumption that forced migration is primarily motivated by negative 'push' factors to emigrate, we found that only 15% of the forced migrants had come to Hungary because of the more favourable economic situation here, and that the majority had been driven by adverse conditions in Romania. 15% mentioned atrocities or discrimination experienced personally, or by someone in their families. 18% said the absence of equal cultural rights for Hungarians was the main reason motivating their decision to move, while 26% mentioned the hopeless economic and political situation in Romania.

As to the unidirectional nature of forced migration, we found that just 3% of forced migrants had plans to return to Romania if the situation there improved.

Given these findings, we can rightfully consider the move of ethnic Hungarian Transylvanians to Hungary to be forced migration rather than migration (section 4). In principle, every migrant in our survey had the chance to plan his move to some extent (section 1).

Based on our findings related to section 3, it seems appropriate to assume that reconversion took a different form for legal forced migrants than it did for illegal forced migrants. Further, on the basis of section 2, forced migrants presumably experienced the post-migration phase as a crisis. They also definitely saw the acquisition of what are everyday belongings for Hungarians to represent an improvement in their standard of living. This can be considered to be the result of a successful reconversion of their pre-migration capital.
 

Well-being in the pre- and post-migration phases

To what extent were forced migrants able to maintain their standards of living? As Table 1 shows, there are no strong correlations between the pre- and post-migration 'well-being' variables, with the exception of the fact that those who had owned a car in Romania tended to be wealthier than their 'less mobile' counterparts.

Table 1. Correlation of material well-being in Transylvania and Hungary (the Pearson correlation)
N = 985

in Hungary	Income	Wealth	Housing
Income in Transylvania	.0828x	.0604	-.0092
Car in Transylvania	-.0288	.3607xx	.0102

Level of significance: ^x = .01 ^xx = .001

The very weak positive correlation suggests that (in the short term) the degree of wealth has changed in the course of migration. I would suggest that a successful conversion and/or reconversion has taken place, but not for everybody.⁵ Table 1 shows that possession of a car was the only physical factor which substantially influenced post-migration wealth.⁶
 

The material capital conversion/reconversion model

As previously indicated, forced migrants can rely on three resources in the post-migration phase: the cultural capital they had accumulated, the direct investments they had made in migration, and their network capital.

We operationalised cultural capital with diplomas, knowledge of foreign languages⁷, the number of books owned, the father's education, and the respondent's happiness in childhood. Each of these factors is likely to help increase post-migration wealth (in the short as well as the long run) either directly (as assets that can be sold in the labour market) or indirectly (ease of communication with officials, appropriate behaviour, self-confidence, etc.).

It is assumed that direct investments made in migration contributed to the success of migration and increased post-migration wealth (diplomas could be smuggled out of Romania with a bribe, for instance). In our sample such resources include, among other things, networks of friends in Hungary, and trips taken to Hungary in the past as well as investments in the form of bribe given either for a passport - in case of legal migration - or for the border guard to 'look the other way' - in case of illegal migration.

Finally, personal networks represent capital just as money does. The only difference is that units in personal networks have 'real faces'. We considered networks to be of particular importance when the individual could foresee and anticipate the coming crisis, and when financial support was unavailable. Being 'intangible', networks play a decisive role in reconversion as they are the kind of resource easiest to accumulate, maintain (perhaps even over several generations) and convert into other types of resources. Three variables characterise the three forms of networks that are considered in our model: personal networks (which may encourage the individual to migrate), earlier visits to Hungary, and Hungarian visitors to Transylvania.

The model compares the level of material well-being in the pre- and post-migration period (Table 4). To describe the channels through which pre-migration material wealth was converted or reconverted in the post-migration period the model characterises the amount of cultural and network capital held by the respondents. In the next step of our analysis we measured the degree of reconversion and the level of direct investment in migration. Finally, in the post-migration phase we measured the role played by formal social support (both governmental and non-governmental). We assumed that the welfare system provided greater support for those in less favourable positions. Consequently, welfare benefits should correlate negatively with all the other categories of capital.

The data on legal and illegal forced migrants were processed separately.

We used two sets of control variables in our model. The two basic socio-demographic variables (sex and age) were used to determine the role played by the different capital types in the reconversion process. Post-migration wealth was determined by adding subjective variables to the model, since we assumed that these can be significant indicators of wealth in general (see the psychosomatic variables in Table 4).

The model was tested with stepwise regression equations. The two material indicators of post-migration well-being (income and 'wealth') were tested separately. We assumed that:

- both indicators of post-migration well-being would be most strongly influenced by the conversion channels used,
- the network capital variables would play an especially crucial role in the process, - the direct transfer of pre-migration wealth would play no significant role in the model, and
- the aforementioned characteristics would differ to some degree between illegal and legal forced migrants.
 

The analysis

The following table illustrates the effect various factors had on the post-migration incomes of legal and illegal forced migrants.

Table 2. Factors affecting the per-capita income of legal and illegal forced migrants (gradual regression with multiple variables)⁸

	Illegal forced migrants
	B	Beta	T-Value	Level of Significance
LATDBERD	2.16777	.25439	4.599	.0000
SZSFTSUM	-2.26075E-03	-.17534	-3.138	.0019
NEM	20.28360	.15242	2.651	0085
(Constant)	35.59704		4.013	.0001
Adjusted R-Square .11389
	Legal forced migrants
SZSFTSUM	-1.22271E-03	-.28141	-6.942	.0000
NEM	6.39187	.15261	3.739	.0002
LOSZA	1.58118	.12632	3.041	.0025
(Constant)	38.51336		6.166	.0000
Adjusted R-Square .13127

Being a male and not applying for welfare support increased income for both types of forced migrants.

The more extensive the personal networks, the higher an illegal forced migrant's income. Human capital produces the same effect for legal forced migrants.

To sum up, what we find among legal forced migrants is a reconversion of wealth through human capital. This is 'normal', and would also take place without migration (more valuable human capital is likely to result in higher incomes practically everywhere in the world).

However, this reconversion follows an 'irregular' pattern for illegal forced migrants. In their cases it is pre-migration network capital that ensures higher post-migration incomes.

When we look at the other components of post-migration material well-being, we find two factors which increase 'wealth' to the same degree for legal and illegal forced migrants: Both welfare support and direct investment into migration (in the form of cash or a car) increase 'wealth' (although to varying degrees). This suggests that post-migration material wealth can also be increased by a reconversion of pre-migration material assets.

Table 3. Factors increasing wealth for legal and illegal forced migrants

	illegal forced migrants
	B	Beta	T-value	Level of Significance
VESZTLEJ	.95216	.26006	4.693	.0000
SZSFTSUM	.93473	.1399	2.497	.0131
LEJHOZLE	.95535	.14234	2.573	.0106
UTMAGYDB	317.90312	.11359	2.041	.0421
(Constant)	9099.44534		3.289	.0011
Adjusted R-Square .12002
	legal forced migrants
AUTOERD	35455.84222	.36398	9.299	.0000
KOR	1360.24375	.14472	3.60	.0003
SZSFTSUM	1.23831	.13562	3.358	.0008
(Constant)	-1497.88917		-.115	.9086
Adjusted R-Square .20237

Direct investment into migration strongly improves illegal forced migrants' chances of gaining wealth. The crucial role played by direct investment in increasing post-migration wealth can be conceptualized as follows: Those with the fmancial resources to pay bribes and fund pre-migration trips to Hungary are also able to increase their wealth in the post-migration phase. This is not due to their personal features (that they are highly mobile, innovative, and show an entrepreneurial spirit) but to the fact that both processes can help reduce the crisis. Probably both bribes and trips made to Hungary are strongly tied to one another and to the black market as migrants can use their trips to smuggle goods into Hungary⁹ (presumably such smuggling is accompanied by bribes to the border guards) and thus convert Romanian lei into Hungarian forints.
 

Conclusions

The assumption that it is impossible to reconvert pre-migration wealth was proven wrong. In legal forced migration a car is the major asset which most strongly determines the level of post-migration wealth - at least in the short term¹⁰.

The assumption that investment into migration, and cultural and network capital are of great importance in reconverting wealth in the course of migration seems to have been supported by our models, as has the hypothesis that there are differences between legal and illegal forced migrants.

Table 4. Definitions of the model variables and the respective values for legal and illegal forced migrants in the sample

	Total	Illegal	Legal forced migrants
Age (average between 19-35)	26	25	27
Sex (1 = male) (%)	60	72	54
Human Capital variables
Formal education (number of grades)	11.1	11.0	11.9
Knowledge of foreign language (1=yes) (%)	35	22	41
Books at home (# of volumes)	291	171	350
Father's education (number of grades)	10.5	9.7	10.8
Happy childhood (1=yes) (%)	72	63	76
re-migration material well-being
Owned a car in Transylvania (1=yes) (%)	25	19	29
Foreign relatives, friends (1=yes) (%)	51	37	57
Previous foreign visits (1=yes) (%)	32	14	41
Network Capital
Encouraged by others to move (1=yes) (%)	18	12	21
Hungarian visitors in 1988 (# of persons)	5.6	4.9	5.9
Number of previous visits to Hungary (#)	7.2	5.1	8.3
Post-migration support from the personal network
Number of support	1.8	1.4	2.0
Post-migration accommodation (1=yes) (%)	63	31	80
Advice on employment (1=yes) (%)	50	49	50
Post-migration institutional support
Number of support	2.6	3.0	2.4
Value of financial support (Ft)	5238	5564	5071
Post-migration psychosomatic state
Great self-confidence(%)*	29	38	25
"Better future" alternative (1=yes) (%)	84	88	81
Afraid of the future (1=yes) (%)	23	17	26
Worried (1=yes) (%)	40	38	41
Gastric pains (1=yes) (%)	14	12	15
Nightmares (1=yes) (%)	32	29	34
Post-migration material well-being
Estimated value of car and durable (Ft)	26282	18284	30378
Per capita average monthly income (Ft)	5447	5418	5467

* Percentage of those classifying themselves under 9 and 10 on a scale of 1 to 10.

LATDBERD: number of Hungarian visitors in Transylvania
SZSFTSUM: the amount of social aid in forints NEM: sex
LOSZA: schooling (by grades passed)
VESZTLEJ: the amount of bribes paid in lei
LEJHOZLE: the amount of Romanian currency brought to Hungary
UTMAGYDB: the number of previous trips to Hungary
ATOERD: possession of a car in Transylvania
 

References

Bourdieu, P., Distinctions. Cambridge, Mass.: Harvard University Press, 1988.

Forró, T.-H. Havas, Arad után - Ki tudja merre? (Leaving Arad - who knows for where?) Budapest: Háttér, 1988

Róbert, P.-E. Sik, "Occupational attainment of Transylvanian refugees in Hungary", in: Workshop: Transformation Processes in Eastem Europe, NWO, SSCW, SRO, Dec. 3-4, 1992. The Hague, pp. 19-46.

Sik, E., "Transylvanian Refugees in Hungary and the Emergence of Policy Networks to Cope with Crisis". Journal of Refugee Studies, vol. 5, 1992, 1.

Sik, E. "Transylvanian Refugees in Hungary". In: Andorka, R. et al. (eds.), Social Report. Budapest: TÁRKI, 1992, pp. 366-378.
 

1. I use the term forced migration (and forced migrant) instead of seeking refuge (and refugee) or migration (and economic migrant) for two reasons: First, Hungary had yet to sign the Geneva Convention in the period covered by the analysis, thus (legally) there were no refugees in Hungary before the autumn of 1989. Second, ethnic Hungarians from socialist Romania took refuge in socialist Hungary for more than economic reasons - at the time Hungarians were highly discriminated against in Romania.

2. In a separate study we examined the role migration plays in status acquisition (Róbert-Sik 1992).

3. It was dangerous for forced legal migrants to take their school certificates across the border, yet they are the only documents which represent the existence of cultural capital. The lack of school certification significantly reduced immigrants' chances of finding work.

4. Empirically, the model has some other characteristics which may provide a unique opportunity to test the conversion phenomenon:

- conversion took place at almost the same time for every respondent - the length of the post-migration phase was almost the same for every forced migrant,
- the macro-economic and political environment was similar for the forced migrants since all of them came from the same country,
- in many respects the population of forced migrants tan be considered homogeneous since they have the same cultural background, are the same age and have at least one similar socio-psychologically significant feature - they all emigrated.

5. Negative correlations would be shown when, for example, refugees from a revolution (who have lost their power) face lower standards of living (for a short period) in their host country, or when an utterly poor population improves its standard of living through migration.

On the other hand, strong positive correlations are found in the case of peaceful transitions when well-to-do forced migrants can take their wealth with them to the host country, while the poor remain poor despite their flight.

6. Another asset equal in its effect was Romanian and Hungarian money. In fact, many migrants brought both Romanian lei and Hungarian forints with them - but only in very small amounts (more to cover their daily expenses in their first few days in Hungary than as a capital asset).

7. For the purposes of our analysis, this represented the knowledge of a language other than Hungarian or Romanian.

8. This table only includes variables that showed significant impacts. A definition of the variables is given in Table 4.

9. This has not been proven, but seems logical. Earlier trips to Hungary may have been helpful in network building since such trips were often planned with the help of personal networks in Hungary.

10. This is partly due to the fact that the first wave of Transylvanian forced migrants came to Hungary to attend the first Formula 1 Race ever held in a communist country (in August,1988). Naturally, being (or pretending to be) car fans, most came to this event by car.