Risk of Economic Violence: A New Quantification

Abstract

This paper defines the first internationally comparable measure of the risk of economic violence to acknowledge its prevalence in different countries and its geographical and gender heterogeneity. Thanks to the availability of micro-data from the OECD/International Network on Financial Education survey, currently used to track financial literacy in different countries, we define a measure of the risk of economic violence (REV) that takes into consideration three macro-areas: (a) the risk of being prevented from acquiring and accumulating financial resources; (b) the risk of being unaware and not having access to personal and/or household financial resources; and (c) the risk of financial dependency. The definition of the new economic violence risk measure (REV) then allows us to verify with real data the presence of women’s greater exposure to the risk of economic violence and the presence of gender differences in the determinants of economic violence risk. Finally, we verify that financial literacy protects individuals from the risk of economic violence, without gender differences.

1. Introduction

Economic violence is conceived as any act that involves making or attempting to make an individual financially dependent by maintaining total control over financial resources, withholding access to money, and/or forbidding attendance at school or employment. At the European level, the European Institute of Gender Equality (EIGE) defines economic violence as “any act or behavior that causes economic harm to an individual” (EIGE 2017). Therefore, economic violence involves behaviors aimed at controlling an individual’s ability to acquire, use, or maintain money, credit, property, or other economic resources, which harms her economic security and potential to achieve self-sufficiency.

Therefore, it often occurs in the context of intimate relationships. In particular, the control of economic resources is one of the main reasons for leaving abusive relationships. Women in heterosexual relationships are disproportionately affected by economic violence (Sharp-Jeffs 2021), so much so that at the European level, economic violence has been defined as a common form of violence against women rooted in gender inequality and reinforced by traditional gender norms (European Commission 2022; EIGE 2024). Indeed, the results of an EU-wide survey on gender-based violence reveal that on average, 12 percent of women in the EU have experienced abuse involving economic violence by a partner (Fundamental Rights Agency—FRA 2014).

The impact of economic violence is multifaceted, affecting not only the immediate financial stability of victims but also their long-term economic opportunities and psychological well-being. Survivors often face significant barriers to achieving financial independence, which perpetuates a cycle of dependency and abuse. The recognition of economic violence as a distinct form of abuse highlights the need for targeted interventions and support systems that address its unique characteristics and consequences. To effectively address the phenomenon, it is necessary to recognize and attempt to quantify the gendered and intersectional nature of economic violence and the increased vulnerability of victims based on factors of race, ethnicity, age, socio-economic status, gender identity, and migrant status.

However, the available data on economic violence, to date collected from ad hoc surveys (such as in Europe the survey conducted by FRA (2014)), was considered not comparable at the international level because data collection practices in different countries are different, many countries do not collect sufficient data on the characteristics of survivors, perpetrators, or their relationships, and mainly because there is a lack of common definitions of economic violence at the international level.

Consequently, we tried to create a measure identifying the risk of being exposed to economic violence based on internationally comparable data so that we could identify its prevalence and possible determinants through an intersectional approach. Indeed, to effectively address the phenomenon of economic violence, it is necessary to recognize its gendered and intersectional nature, which implies an increased vulnerability based on different factors such as gender identity, age, education levels, immigration status, socio-economic status, family care responsibilities, and financial education levels.

Thanks to the availability of the microdata of the extensive OECD/International Network on Financial Education (INFE) survey, currently used to analyze financial literacy, we define a composite measure of the risk of economic violence—REV—which takes value 1 when an individual is at risk of economic violence and 0 otherwise. The REV defines an internationally comparable measure to define its prevalence in different countries and its distribution across the socio-demographic spectrum.

We then used the measure of economic violence risk (REV) created to go and test, with real data, three relevant hypotheses:

(a)

Women are more at risk of economic violence;

(b)

There are gender differences in the determinants of economic violence risk;

(c)

Higher levels of financial literacy protect against the risk of economic violence.

The results of our analysis allow us to validate all three of the above hypotheses and go on to suggest policy directions.

The paper is structured as follows: Section 2 defines the complexity of economic violence and presents its main dimensions based on the relevant literature in the field. Section 3 proposes a measure of the risk of economic violence based on the literature review of Section 2 and the existing data from OECD/INFE surveys. Section 4 analyzes the main determinants of the risk of economic violence from a gender perspective. The last section proposes some policy recommendations and new research directions needed to broaden the understanding of the phenomenon of economic violence.

2. Economic Violence: Main Dimensions

Following the reference literature, we identified three main dimensions of economic violence that appear to be recurrent within the very fragmented work on the topic, which helped us in constructing our measure of the risk of economic violence (next section).

2.1. Prevention of Resource Acquisition and Accumulation

The first—and most common—group of economic violence tactics falls under the big umbrella of the prevention of resource acquisition and accumulation. This comprehends all those tricks devoted to impairing the individual’s present and future possibility of having the means to sustain herself and to construct a healthy financial situation. The literature refers to different attitudes, such as the basic suppression of any possible financial goals, the more concrete impediment of participation in the active labor force, or even the long-term prevention in the form of retirements and savings sabotage. In particular, many studies analyze the act of preventing woman’s ability to obtain and maintain employment outside the home to acquire their resources, perpetuated by abusive men (Aguilar and Nightingale 1994; Brewster 2003; Curcio 1997; Hudson and McIntosh 1981; Sable et al. 1999; Shepard and Pence 1988; Tolman and Wang 2005; Von De Linde 2002; Walker 1979).

2.2. Prevention of Awareness and Use of Personal/Household Financial Resources

The second recurrent macro-area of economic abuse involves preventing individuals from being aware of and using (personal and household) financial resources. Specifically, in the intimate partner violence (IPV) literature, abusive men exercise power by controlling how resources are distributed and by monitoring how they are used (Anderson et al. 2003; Brewster 2003; Davies and Lyon 1998; Dobash and Dobash 1979; Hofeller 1982; Martin 1976). Again, this aspect of the phenomenon encompasses various forms, from the prevention of awareness of financial assets—like household income—and of financial activities to the cruder absence of even basic financial instruments like a current account, debit card, mobile payment card, or payment card. As said, this form of prevention also manifests in the realm of awareness. In fact, not only economic control occurs when the perpetrator prevents a person from having access to financial resources (Anderson et al. 2003; Wettersten et al. 2004; Postmus et al. 2015; Sanders 2015), but also when a perpetrator may lie about shared properties and assets (Brewster 2003; Von De Linde 2002).

2.3. Financial Dependency

Finally, the third dimension that can be identified is the ultimate creation of financial dependency. This macro-area groups all those measures devoted to ensuring that the violence is long-term and that the target has no escape. In addition to dictating and monitoring how resources are used, some batterers intentionally deplete women’s available resources as a means of limiting their options. This can occur in a variety of ways, including stealing money or financial resources, creating costs, and generating debt (Sharp-Jeffs 2021).

3. Risk of Economic Violence: A Quantification

Based on the three main dimensions described above, we create a composite measure of the risk of economic violence that could be comparable at the international level, based on the questions asked in the framework of the OECD/INFE survey. There are several advantages to using OECD/INFE data in the measurement of issues related to economic violence. First, the extensive nature of the dataset allows for international comparisons. Moreover, in the pursuit of achieving a robust sample, the OECD suggested commissioning institutions discuss with survey agencies the benefits of setting quotas and/or including booster samples of hard-to-reach groups and the implications in terms of sample size and confidence in the results. Second, we define situations of risk, reducing the problem of different cultural and social biases in the self-identification of being at risk of economic violence and differences in the definition of economic violence in different countries. Lastly, by nature, the dataset focuses on individuals and not on households as a unit of analysis. In fact, in introducing the questionnaire, it is made clear to respondents that one is interested in his/her situation and opinions rather than those of the household or main income earner, unless otherwise indicated.

Of course, there are also many limitations to using the data collected within the INFE by the OECD. First and foremost, the survey is not designed to collect evidence of direct lived experience of economic violence; consequently, the topic is examined indirectly by identifying the ‘risk’, i.e., the susceptibility of individuals to experience economic violence. However, although these factors may be correlated with the incidence of economic violence, they are neither necessary nor sufficient for economic violence to occur, but only a measure of the potential risk of being targeted by it. Furthermore, as the survey is not dedicated to the topic of violence, some scenarios of economic violence are not specifically captured by the survey questions used to construct our ‘Risk of Economic Violence’ (REV) measure. Another element to consider is that this survey was designed specifically to consider levels of financial literacy. This is reflected in the scant consideration of the dynamics of intra-household bargaining, power relations, and violence that the literature usually refers to in the context of economic violence.

However, we believe that the data from the INFE/OECD survey provides a good basis for quantifying the phenomenon of economic violence to provide a possible comparison over time (the survey is conducted regularly) and between countries.

To identify if an individual is at risk of economic violence, we used the OECD/INFE survey data collected throughout 2019 and the first quarter of 2020, before the effects of the COVID-19 pandemic. More specifically, the dataset includes information from the countries that agreed to share upon request their data for research purposes, and it contains 21,324 responses from 19 participating countries: Austria, Bulgaria, Colombia, Croatia, the Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Montenegro, North Macedonia, Peru, Poland, Romania, and Slovenia. It is worth noting that there is an unbalanced geographical coverage (see

Table 1. Country list and sample sizes.

Countries	Number of Participants	Share of Women (%)
Austria	1418	52.89
Bulgaria	1047	52.14
Colombia	1200	51.33
Croatia	1079	55.70
Czech Republic	1003	50.75
Estonia	1005	55.42
Georgia	1056	63.73
Honk Kong, China	1002	54.69
Hungary	1001	52.65
Indonesia	1000	49.00
Italy	2036	49.95
Malta	1013	51.23
Moldova	1074	57.82
Montenegro	1030	51.55
North Macedonia	1076	50.74
Peru	1205	49.96
Poland	1000	51.40
Romania	1060	55.75
Slovenia	1019	52.21
Total *	21,324	52.94
OECD (8) total **	9682	45.40
EU total	12,681	59.47

* Total refers to the sum of the data from the 19 countries analyzed: Austria, Bulgaria, Colombia, Croatia, Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Monte-negro, North Macedonia, Peru, Poland, Romania, and Slovenia. ** The OECD member countries in the sample are Austria, Colombia, the Czech Republic, Estonia, Hungary, Italy, Poland, and Slovenia. Source: authors’ elaborations, with data from OECD INFE (2020).

): the European continent counts for 11 out of 19 countries (over 59.4% of the total sample). This highlights the evident lack of representation for Africa and other major developed countries like the United States and the United Kingdom.

Using OECD/INFE data, we compute a measurement tool that synthesizes the complexity of economic violence as a multi-dimensional concept into a user-friendly and easily interpretable measure. We called this measure REV—Risk of Economic Violence. It takes a value of 1 whenever an individual is at risk of economic violence and 0 otherwise. The REV consists of three core domains, identified following the literature review in Section 2 as being at risk of experiencing the main dimensions of economic violence: prevention of resource acquisition and accumulation, prevention of awareness and use of personal/household financial resources, and financial dependency. The REV is computed as follows: First, three variables within each domain are combined to obtain domain indices. Next, domain indices are combined to get the overall REV. More specifically, the REV, assessing whether a respondent is at risk of economic violence, is by construction equal to 1 whenever at least two of the three domains are equal to 1. In turn, each of the three domains is equal to 1 if at least two of the three variables used to explain them are equal to 1 themselves.

Table 2. Definition of variables for the measurement of REV.

Variable Used	Description
Domain 1: at the risk of being prevented from acquiring and accumulating resources
Inactive	=1 if the responder’s current work situation is best described either as “looking after the home ” or “not working and not looking for work” (QD10)
No financial goals	=1 if the responder has no financial goals (such as paying university fees, buying a car, or becoming debt-free) or answers “Don’t know” when asked about them (QF5)
No income	=1 if the responder states that (personally) her/his income has not quite covered their living expenses in the last 12 months or answers “Don’t know” when asked about it(QF11)
Domain 2: at risk of being unaware of and not having access to personal/household financial resources
Unawareness of personal financial affairs	=1 if the responder strongly disagrees or answers “Don’t know” with the affirmation “I keep a close personal watch on my financial affairs” (QS1)
No financial instrument	=1 if the responder does not hold a current account debit card mobile payment facility or payment card
No financial savings investments	=1 if the responder does not hold a savings, investment, or retirement product
Domain 3: at risk of financial dependency
Impossibility to survive a financial shock	=1 if the responder answers “No” or “Don’t know” when asked if she/he (personally) would you be able to face an expenditure shock without borrowing the money or asking family or friends to help (QF4)
No decisions	=1 if the responder answers “someone else makes these decisions” or “Don’t know” to the question “who is responsible for making day-to-day decisions about money in the household” (QF1)
No plans for retirement	=1 if the responder answers “5 not at all confident” or “Don’t know” to the question on the confidence in having done a good job of making financial plans for her/his retirement (QF8)

Source: authors’ elaborations, with data from OECD INFE (2020).

presents an overview of the variables concurring in each domain. For the overall REV, we performed a χ² test to confirm whether there are significant differences in the distribution of men and women.

3.1. Domain 1: At the Risk of Being Prevented from Acquiring and Accumulating Resources

The first domain, displayed in

Table 3. Individuals at risk of being prevented from acquiring and accumulating resources.

	At Risk of Being Prevented from Acquiring and Accumulating Resources. (%)			A, Inactive in Formal Labour Market (%)			B, No Financial Goals (%)			C, No Income (%)
	Tot	Women	Men	Tot	Women	Men	Tot	Women	Men	Tot	Women	Men
Austria	9.66	10.4	8.83	3.17	5.87	0.15	49.79	46.93	52.99	14.03	14.13	13.92
Bulgaria	32.28	32.23	32.34	2.2	3.66	0.6	52.91	51.28	54.69	52.34	53.85	50.7
Colombia	36.08	38.96	33.05	7.83	4.83	11.47	58.92	59.74	58.05	56.08	61.04	50.86
Croatia	22.06	22.8	21.13	4.54	7.49	0.84	54.12	52.08	56.69	32.72	34.78	30.13
Czech Republic	12.36	11.79	12.96	2.19	4.13	0.2	58.92	58.35	59.51	18.84	18.66	19.03
Estonia	18.41	18.31	18.53	3.08	4.85	0.89	52.44	53.68	50.59	32.94	31.42	34.82
Georgia	47.16	53.94	35.25	20.17	30.01	2.87	63.16	65.97	58.22	56.25	57.36	54.31
Hong Kong, China	18.16	25.73	9.03	18.36	33.58	-	37.92	41.79	33.26	20.36	21.35	19.16
Hungary	15.58	17.65	13.29	2.3	4.36	-	55.94	57.12	54.64	21.08	23.34	18.57
Indonesia	24.2	42.24	6.86	29.3	55.71	3.92	6.8	7.96	5.69	64.3	64.29	64.31
Italy	3.73	7.28	0.2	12.13	23.89	0.39	-	-	-	28.83	30.38	27.28
Malta	21.62	28.32	14.57	7.4	12.33	2.23	55.48	56.45	54.45	34.25	37.76	30.57
North Macedonia	28.07	35.71	20.19	10.96	18.86	2.26	57.71	58.97	56.42	36.99	42.12	31.7
Moldova	40.69	46.38	32.89	12.38	17.07	5.96	52.79	55.56	49.01	63.22	66.51	58.72
Montenegro	30.39	34.09	26.45	7.09	11.3	2.61	52.33	52.17	52.51	51.46	55.93	46.69
Peru	40.58	47.51	33.67	14.94	26.91	2.99	55.35	55.48	55.22	62.24	65.28	59.2
Poland	9.3	10.31	8.23	1.8	2.72	0.82	58.4	59.92	56.79	15.3	14.59	16.05
Romania	26.79	29.78	23.03	5.09	7.95	1.49	53.02	54.65	50.96	45.38	46.36	44.14
Slovenia	0.79	1.13	0.41	1.86	3.2	0.41	-	-	-	27.77	31.95	23.2
Total *	22.29	26.6	17.45	8.87	14.9	2.08	44.3	45.39	43.07	38.23	40.33	35.88
OECD (8) average **	12.52	14.06	10.86	5.15	8.28	1.78	37.96	38.33	37.55	27.12	28.45	25.69
EU average	14.65	16.54	12.56	4.78	8.48	0.68	41.24	40.94	41.51	29.04	30.41	27.52

* Total refers to the sum of the data from the 19 countries analyzed: Austria, Bulgaria, Colombia, Croatia, Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Monte-negro, North Macedonia, Peru, Poland, Romania, and Slovenia. ** The OECD member countries in the sample are Austria, Colombia, the Czech Republic, Estonia, Hungary, Italy, Poland, and Slovenia. Source: authors’ elaborations, with data from OECD INFE (2020).

, explores the risk of being prevented from acquiring and accumulating resources. Given its purpose, it is composed, respectively, of a measure of inactivity in the formal labor markets, a measure of the absence of basic financial goals; a measure assessing the poor state of the respondent’s income (i.e., if the respondent states that her/his income has not quite covered their living expenses in the last 12 months). With this domain, we identify people with no income and no possibility to generate income due to the absence of access to the paid labor market, as well as those who have no financial objectives and hence do not think of emancipating in the long-term. It is interesting to note that in our sample, 17.45% of men and 26.6% of women are at risk of being prevented from acquiring and accumulating resources. These gender differences persist if only the EU or OECD countries are considered, even if the shares are lower. Moreover, while for the second and third sub-components—i.e., ‘no financial goals’ and ‘no income’—there are gendered differences, but somewhat muted in their intensity, the first sub-component measuring the share of people ‘inactive in the formal labor market’ reflects the traditional gender division of labor that sees women more than men outside the paid labor force.

3.2. Domain 2: At Risk of Being Unaware of and Not Having Access to Personal/Household Financial Resources

The second domain, in

Table 4. Individuals at risk of being unaware of and not having access to personal/household financial resources.

	At the Risk of Being Unaware of and Not Having Access to Personal/Household Financial Resources (%)			A. Unawareness of Personal Financial Affairs (%)			B. Not Being the Holder of a Bank Account, Debit Card, Mobile Payment Card or Payment Card (%)			C. No Financial Saving Investments (%)
	Tot	Women	Men	Tot	Women	Men	Tot	Women	Men	Tot	Women	Men
Austria	1.83	2.27	1.35	1.76	2.13	1.35	1.76	2.4	1.05	10.72	10.53	10.93
Bulgaria	44.13	44.32	43.91	3.25	2.93	3.59	74.98	74.73	75.25	49.76	50.18	49.3
Colombia	58.5	63.96	52.74	15.08	18.18	11.82	76.25	78.25	74.14	63.83	68.34	59.08
Croatia	10.57	9.82	11.51	7.23	6.49	8.16	49.58	48.42	51.05	7.88	7.82	7.95
Czech Republic	8.28	7.47	9.11	3.39	3.14	3.64	9.67	9.23	10.12	33	32.81	33.2
Estonia	20.4	20.11	20.76	3.58	2.33	5.13	23.08	23.34	22.77	61.89	62.12	61.61
Georgia	52.84	54.38	50.13	6.63	6.24	7.31	78.6	80.68	74.93	55.68	56.61	54.05
Hong Kong. China	-	-	-	0.4	0.18	0.66	-	-	-	-	-	-
Hungary	33.17	34.72	31.43	12.59	11.95	13.29	28.97	31.31	26.37	76.12	76.66	75.53
Indonesia	14	16.94	11.18	1.4	1.63	1.18	62.5	64.29	60.78	16.3	18.98	13.73
Italy	32.96	35.4	30.52	8.6	8.36	8.83	34.87	37.76	31.99	78	79.94	76.05
Malta	-	-	-	2.76	1.73	3.85	-	-	-	-	-	-
North Macedonia	26.49	29.3	23.58	5.11	5.49	4.72	62.17	61.9	62.45	27.42	30.4	24.34
Moldova	69.27	69.24	69.32	13.04	12.24	14.13	85.85	86.15	85.43	74.95	76.49	72.85
Montenegro	29.03	30.32	27.66	13.2	11.49	15.03	58.54	58	59.12	27.09	29.76	24.25
Peru	40.17	45.02	35.32	14.19	15.45	12.94	59.75	65.61	53.9	36.6	40.2	33
Poland	11.1	10.31	11.93	6.6	7.39	5.76	8	6.42	9.67	62.6	62.45	62.76
Romania	33.11	32.83	33.48	3.68	3.21	4.26	61.89	61.08	62.9	38.58	40.27	36.46
Slovenia	0.39	0.19	0.62	0.59	0.38	0.82	-	-	-	75.27	79.89	70.23
Total *	26.13	28.67	24.39	6.65	6.55	6.77	40.77	42.1	39.28	43.14	44.73	41.36
OECD (8)	22.04	23.06	20.94	6.7	6.52	6.87	24.26	25.07	23.29	57.98	59.26	56.61
Average **
EU average	18.6	18.9	18.28	5.1	4.74	5.5	26.89	27.57	26.14	43.14	46.74	45.68

* Total refers to the sum of the data from the 19 countries analyzed: Austria, Bulgaria, Colombia, Croatia, Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Monte-negro, North Macedonia, Peru, Poland, Romania, and Slovenia. ** The OECD member countries in the sample are Austria, Colombia, the Czech Republic, Estonia, Hungary, Italy, Poland, and Slovenia. Source: authors’ elaborations, with data from OECD INFE (2020).

, treats the impossibility of using and being aware of personal/household financial resources. In this case, both the unawareness aspect is explored by monitoring a question on the perception of the responders on the statement “I keep a close personal watch on my financial affairs”; and the two remaining variables check whether the responder does not hold a bank account, debit card, mobile payment facility, or payment card, and whether the responder does not hold a savings, investment, or retirement product. Compared to the previous, this domain presents milder gender differences; a possible explanation for this can be found in the hypothesis that those who already have access to the resources are the ones that already gave some type of reasoning on the financial resources and hence have more financial skills (and less fragility). Once again, there is the incorporation—even if to a very limited extent—of a more long-term perspective by adding the retirement savings, which has a small, minor contribution, but it is important when talking about emancipation and gender gaps in poverty (Corsi et al. 2016).

3.3. Domain 3: At Risk of Financial Dependency

The third and final domain, in

Table 5. Individuals at risk of financial dependency.

	At Risk of Financial Dependency (%)			A, Impossibility to Survive to Financial Shock (%)			B, No Decisions (%)			C, Retirement Planning (%)
	Total	Women	Men	Total	Women	Men	Total	Women	Men	Total	Women	Men
Austria	3.88	3.87	3.89	21.58	23.2	19.76	4.3	4.13	4.49	4.02	3.87	4.19
Bulgaria	23.3	23.26	23.35	53.77	54.4	53.09	3.92	4.03	3.79	31.9	31.68	32.14
Colombia	29.42	35.71	22.77	58.58	61.53	55.48	29.75	32.47	26.88	21.83	28.41	14.9
Croatia	15.01	17.97	11.3	56.16	60.23	51.05	3.52	4.16	2.72	19.28	21.13	16.95
Czech Republic	10.87	11.59	10.12	35.89	35.95	35.83	6.38	4.91	7.89	14.26	16.9	11.54
Estonia	22.29	23.52	20.76	46.67	49.91	42.63	4.68	3.41	6.25	32.94	34.11	31.47
Georgia	31.63	33.28	28.72	73.2	74.29	71.28	9.94	9.96	9.92	31.06	32.54	28.46
Hong Kong, China	9.78	11.5	7.71	22.26	25.36	18.5	12.97	16.06	9.25	6.59	7.48	5.51
Hungary	16.58	15.94	17.3	47.45	48.96	45.78	6.19	5.88	6.54	22.68	22.58	22.78
Indonesia	3.8	3.88	3.73	44.9	44.69	45.1	1.4	1.43	1.37	3.8	3.88	3.73
Italy	19.35	21.83	16.88	33.06	33.82	32.29	20.68	21.93	19.43	23.97	26.65	21.3
Malta	8.49	10.4	6.48	25.96	27.36	24.49	14.91	15.99	13.77	6.81	8.86	4.66
North Macedonia	16.45	19.6	13.21	52.14	55.13	49.06	11.8	15.02	8.49	15.71	18.68	12.64
Moldova	25.05	27.05	22.3	62.38	68.44	54.08	3.35	3.06	3.75	34.17	34.3	34
Montenegro	26.02	26.74	25.25	58.83	61.39	56.11	13.69	11.68	15.83	30.97	33.52	28.26
Peru	26.39	30.23	22.55	51.29	54.32	48.26	33.53	37.87	29.19	17.93	21.43	14.43
Poland	10.7	9.73	11.73	40.7	42.22	39.09	4.4	2.53	6.38	19.2	17.7	20.78
Romania	11.98	11	13.22	52.55	52.28	52.88	2.92	3.05	2.77	16.89	16.58	17.27
Slovenia	6.97	8.65	5.13	32.38	34.59	29.98	5.3	7.71	2.67	8.64	10.71	6.37
Total *	16.88	18.6	14.95	45.08	47.51	42.34	10.92	11.37	10.4	19.14	20.93	17.12
OECD (8)	15.28	16.75	13.69	38.45	40.16	36.61	11.46	11.61	11.31	18.47	20.27	16.52
Average **
EU average	13.76	14.63	12.79	39.5	41.24	37.57	8	7.97	8.03	18.26	19.32	17.1

* Total refers to the sum of the data from the 19 countries analyzed: Austria, Bulgaria, Colombia, Croatia, Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Monte-negro, North Macedonia, Peru, Poland, Romania, and Slovenia. ** The OECD member countries in the sample are Austria, Colombia, the Czech Republic, Estonia, Hungary, Italy, Poland, and Slovenia. Source: authors’ elaborations, with data from OECD INFE (2020).

, accounts for financial dependency. It comprehends again three parts: the impossibility of surviving an unexpected financial shock (generally identified with the term “financial resilience” by the OECD); the delegation of financial decision making; and the absence of interest in financial retirement plans. Like in the two previous cases, there are also significant gendered differences in this case that expose women to a higher risk of financial violence.

3.4. Risk of Economic Violence: The Computation of the REV Measure at the International Level

Once the three domains have been calculated, we can define the measure of the risk of economic violence (REV) and determine the differences between the countries considered and the gender differences in exposure to the potential risk of being a target of economic violence.

Table 6. Risk of Economic Violence (REV).

	Risk of Economic Violence—REV 1
	Total	Women	Men
Austria	1.83	2.00	1.65
Bulgaria	29.70	29.67	29.74
Colombia	40.17	47.73	32.19
Croatia	8.62	9.48	7.53
Czech Republic	6.78	6.88	6.68
Estonia	13.53	14.00	12.95
Georgia	44.13	49.33	34.99
Hong Kong, China	4.79	5.47	3.96
Hungary	16.48	17.65	15.19
Indonesia	6.90	11.22	2.75
Italy	11.35	15.44	7.26
Malta	1.97	2.70	1.21
North Macedonia	16.91	20.15	13.58
Moldova	42.46	46.70	36.64
Montenegro	23.20	24.67	21.64
Peru	32.86	40.37	25.37
Poland	5.80	5.25	6.38
Romania	18.96	19.46	18.34
Slovenia	0.29	0.19	0.41
Total 2	17.12	19.83	14.06
OECD (8) average 3	12.07	13.94	10.06
EU average	10.35	11.32	9.27
Test χ2 H0: distribution of men = distribution of women		0.00000011 ***

¹ REV indicates the share of individuals at risk of economic violence. ² Total refers to the sum of the data from the 19 countries analyzed: Austria, Bulgaria, Colombia, Croatia, the Czech Republic, Estonia, Georgia, Hong Kong (China), Hungary, Indonesia, Italy, Malta, Moldova, Monte-negro, North Macedonia, Peru, Poland, Romania, and Slovenia. ³ The OECD member countries in the sample are Austria, Colombia, the Czech Republic, Estonia, Hungary, Italy, Poland, and Slovenia. Source: authors’ elaborations, with data from OECD INFE (2020). Asterisks denote levels of significance: *** ≤1%. Source: authors’ elaborations, with data from OECD INFE (2020).

presents the final synthetic REV measure for each country considered in the study.

The results indicate that the shares of individuals at risk of economic and financial violence vary significantly across the 19 countries analyzed. The overall percentage of individuals at risk stands at 17.12%, with a pronounced and significant gender disparity: 19.83% of women are at risk compared to 14.06% of men. This gender gap underscores the differential impact of economic violence, which disproportionately affects women.

Detailed country-specific data reveal substantial variations. For instance, in Georgia, the highest risk levels are observed with 44.13% of individuals affected, and there is a stark gender difference where 49.33% of women compared to 34.99% of men are at risk. Similarly, Moldova shows a high prevalence, with 42.46% of the population at risk (46.70% of women and 36.64% of men). These figures are indicative of significant socio-economic challenges and gender inequities prevalent in these countries. These high percentages suggest a deep-rooted issue within the social and economic backgrounds of these nations, where traditional gender roles and limited economic opportunities for women exacerbate their vulnerability to economic violence.

In contrast, countries like Slovenia and Austria exhibit much lower risk levels. Slovenia reports the lowest risk at just 0.29% (0.19% for women and 0.41% for men), highlighting a potentially more effective socio-economic structure and financial education system. Austria follows with 1.83% (2.00% for women and 1.65% for men). The lower percentages in these countries could be attributed to stronger social safety nets, better gender equality policies, and more widespread financial literacy programs.

Interestingly, the data reveal that developing countries tend to have higher percentages of individuals at risk. For instance, Colombia has a notable 40.17% of its population at risk, with women (47.73%) significantly more affected than men (32.19%). Peru also shows a high risk at 32.86% (40.37% for women and 25.37% for men). These high-risk percentages in developing countries can be attributed to weaker financial infrastructure, less effective financial education programs, and greater gender inequalities. The economic instability and lack of social protections in these regions likely contribute to the elevated risk levels, making it difficult for individuals, especially women, to achieve financial independence and security.

Conversely, the average values for OECD and EU countries are lower than the overall average. The OECD average stands at 12.07% (13.94% for women and 10.06% for men), while the EU average is even lower at 10.35% (11.32% for women and 9.27% for men). These lower averages can be linked to stronger economic systems, more comprehensive social safety nets, and better financial literacy and inclusion policies. The relatively lower percentages in OECD and EU countries suggest that these regions benefit from more robust institutional frameworks and policies aimed at promoting gender equality and financial well-being. As for European countries, EU member states that have ratified the Council of Europe Convention on Preventing and Combating Violence Against Women and Domestic Violence (Istanbul Convention) are obliged to adopt legislative measures to prevent and combat economic violence (Council of Europe 2011). However, in 2021, only nine EU member states explicitly condemned forms of economic violence in their domestic violence laws, including Slovenia, which in our analysis shows the lowest levels of the REV measure (European Commission 2021).

The gender discrepancies highlighted in the REV measure call for a deeper exploration. As a result, the next chapter will go into the details of data analysis to better define the consistency of gender differences and whether there are differences in the definition of factors that may affect exposure to economic violence risk between men and women.

4. Risk of Economic Violence: Main Determinants

Once defined as a measure of the risk of economic violence (REV), we use it to test three hypotheses with real data:

(a)

women are at higher risk of economic violence;

(b)

there are gender differences in the determinants of being at risk of economic violence;

(c)

higher levels of financial literacy can protect people from being at risk of economic violence.

Given the binary nature of the REV, to test these hypotheses, we used a probit model. Recalling that in the probit model, the probability p_i of being at risk or not at risk of economic violence can be expressed as:

$$p_i=Prob\left(Y_i=1|X\right)={\int }_{-\infty }^{{\displaystyle x_i^{\prime }}\beta }{\left(2\pi \right)}^{-1/2}exp\left(-{\displaystyle \frac{t^2}{2}}\right)dt=\Phi \left(x_i^{\prime }\beta \right)$$

(1)

where Φ is the cumulative distribution function of a standard normal variable that ensures 0 ≤ p_i ≤ 1, x is a vector of factors that determine or explain the variation in being at risk, and β is a vector of parameters or coefficients that reflect the effect of changes in x on the probability of being at risk of economic violence. Indeed, although financial dependence influences the risk of economic violence, the literature suggests that other factors such as gender, age, and migration status may also influence financial dependence, and, thus, the resulting greater exposure to the risk of economic violence. In defining the factors to be included in our model, we were guided by the relevant literature. Age appears to be an important factor to consider in defining exposure to the risk of economic violence. Miškulin’s (2020) study of the lifetime prevalence of economic violence in Croatia found that women over the age of 30 were more likely than younger women to have experienced economic abuse. However, an analysis in Germany, contrary to Hungarian findings, found a similar prevalence of economic violence across age groups (Stöckl and Penhale 2015). Another essential factor to analyze in defining exposure to economic violence risk is education level: based on a population study in Australia that included more than 13,000 women, Kutin et al. (2017) found that women with lower levels of education were more likely to experience economic abuse. Migrant status was also found to be associated with higher odds of experiencing economic violence. Bettio and Ticci (2017) estimate, based on their analysis of FRA survey data, that women in the EU who identify as belonging to ethnic or religious minorities are about twice as likely to experience economic violence as women who identify as belonging to majority ethnic and religious groups. Research in Spain suggests that migrant women’s dependence on their partners may be greater than that of non-migrant Spanish women, as migration is associated with job insecurity and more disadvantaged economic situations (Briones-Vozmediano et al. 2014). Also in Spain, an analysis of violence against women conducted in 2019 showed that migrant women are twice as likely to experience economic violence as native women (Ministerio de Igualdad 2023).

In general, the literature underscores the importance of participation in the paid labor market for economic independence and, consequently, for financial independence: lack of remuneration, as well as an increased burden of unpaid care work in the household, tend to increase economic vulnerability and economic/financial subordination. Bettio and Ticci’s (2017) analysis of data from the 2012 FRA survey shows that, across the EU, women who do not work or who work but earn less than their partners are at greater risk of experiencing economic control/sabotage (as indicated by the inability to make decisions about family finances, shop independently, or work outside the home). The literature on maternity income penalties highlights parenting as a factor that exacerbates gender inequalities in pay and earnings (Budig et al. 2012, 2016; OECD 2012). In all OECD countries, the gender pay gap widens significantly during the childbearing and child-rearing years (OECD 2012). Consequently, in our analysis, we also consider family status and the presence of children under the age of eighteen living together. An additional variable of interest in defining exposure to the risk of economic violence is the level of financial literacy, which can increase financial resilience, the ability to withstand and recover from financial shocks, and subsequent long-term financial well-being (OECD 2021). Many studies have shown a consistent gender gap in levels of financial literacy, with men on average more financially literate than women (Bucher-Koenen and Lusardi 2011; Bottazzi and Lusardi 2021; European Commission and Flash Eurobarometer 2023). Consequently, we supplemented our analysis with a variable that is able to define the level of literacy of individuals that is computed by the OECD, and is comparable among the countries considered.

In practice, the elements of vector x that we included in the model can be grouped into three macro areas (see

Table 7. Definition of variables in the Probit model.

Variable Used	Description
At the risk of financial violence	1 = if the responder is likely to be a target of economic violence.
Personal characteristics
Age	of the responder (in one of the following age bands: 18–29; 30–39; 40–49; 50–59; 60–69; 70–79)
Education Level	Highest level of education completed by the responder (no formal education; primary school; lower secondary school or high school; upper secondary school or high school; university-level; post-graduate or equivalent)
Migrant status	=1 if the responder was born in a different country/region than the one from which the questionnaire has been administered
Urban	Size of the community in which the respondent is living on the day of the interview (village with <3000 people; small town with 3000–5000 people; town with 15,000–100,000 people; city with 100,000–1,000,000 people; large city with >1,000,000 people)
Household activities
Married (living with a spouse or partner) and outside the formal labour market	1 = responder has a partner or a spouse in her/his household and she/he is inactive
Parent	1 = responder has at least one child (son/daughter of the respondent or of her cohabiting partner) in her household.
Children in the household	No. of children under 18 in the household
Financial Literacy
FL score	OECD INFE 2020 Financial Literacy score (0–21)

Source: authors’ elaborations, with data from OECD INFE (2020).

):

• personal characteristics that include: (i) sex; (ii) age; (iii) education level; (iv) migrant status; (v) nature of the external living context (i.e., whether the individual lives in a small or big city;
• family care burden that indicates the role within the household, such as (i) family status intersected with the inactivity in the paid labor market; (ii) parenthood status; and (iii) the number of children under 18 years old that cohabit in the household. Other characteristics strictly concerning the specific intra-household bargaining power dynamics were not included due to a lack of data;
• financial literacy. To define the level of financial literacy, we use the Financial Literacy (FL) score proposed and computed by the OECD-INFE, which ranges between 1 and 21. The score is calculated as the sum of three components: (i) knowledge, (ii) behavior, and (iii) attitude (for details, see OECD 2023).

Table 8. Determinants of the probability of being at risk of economic/financial violence, total population.

	A	B	C	D	E
Personal characteristics
Women	0.0316 ***	0.0268 ***	0.0213 ***	0.0251 ***	0.0204 ***
	(0.00545)	(0.00500)	(0.00570)	(0.00806)	(0.00770)
Age, 18–29 years	−0.0514 ***	−0.0735 ***	−0.0914 ***	−0.0734 ***	−0.0460 *
	(0.0163)	(0.0177)	(0.0200)	(0.0269)	(0.0235)
Age, 30–39 years	−0.0852 ***	−0.115 ***	−0.126 ***	−0.0923 ***	−0.0502 **
	(0.0161)	(0.0172)	(0.0197)	(0.0267)	(0.0235)
Age, 40–49 years	−0.0887 ***	−0.112 ***	−0.126 ***	−0.0878 ***	−0.0415 *
	(0.0159)	(0.0172)	(0.0196)	(0.0266)	(0.0235)
Age, 50–59 years	−0.0615 ***	−0.0971 ***	−0.111 ***	−0.0679 **	−0.0189
	(0.0162)	(0.0173)	(0.0198)	(0.0266)	(0.0235)
Age, 60–69 years	−0.0295 *	−0.0837 ***	−0.0917 ***	−0.0362	0.00138
	(0.0167)	(0.0174)	(0.0199)	(0.0270)	(0.0239)
Age, 70–79 years	−0.0152	−0.0723 ***	−0.0697 ***	−0.00331	0.0176
	(0.0181)	(0.0183)	(0.0211)	(0.0289)	(0.0256)
Education level, primary school	−0.162 ***	−0.123 ***	−0.133 ***	−0.158 ***	−0.0309
	(0.0423)	(0.0364)	(0.0427)	(0.0547)	(0.0481)
Education level, lower secondary school	−0.231 ***	−0.178 ***	−0.196 ***	−0.206 ***	−0.043
	(0.0417)	(0.0360)	(0.0425)	(0.0540)	(0.0478)
Education level, upper secondary or high school	−0.233 ***	−0.248 ***	−0.270 ***	−0.299 ***	−0.0905 *
	(0.0415)	(0.0355)	(0.0418)	(0.0527)	(0.0468)
Education level, university	−0.292 ***	−0.322 ***	−0.351 ***	−0.406 ***	−0.162 ***
	(0.0417)	(0.0356)	(0.0419)	(0.0530)	(0.0474)
Education, post-graduate	−0.332 ***	−0.354 ***	−0.384 ***	−0.456 ***	−0.206 ***
	(0.0420)	(0.0359)	(0.0423)	(0.0535)	(0.0485)
Migrant	0.0221	0.0264 *	0.0318 **	0.0380 *	0.0308
	(0.0151)	(0.0140)	(0.0157)	(0.0201)	(0.0190)
Urban, small-town			−0.0378 ***	−0.0386 ***	−0.0429 ***
			(0.00914)	(0.0144)	(0.0134)
Urban, town			−0.0266 ***	−0.0326 ***	−0.0250 **
			(0.00841)	(0.0115)	(0.0109)
Urban, city			−0.0307 ***	−0.0373 ***	−0.0289 ***
			(0.00801)	(0.0108)	(0.0102)
Urban, large city			−0.0271 **	−0.0463 ***	−0.0285 *
			(0.0116)	(0.0161)	(0.0157)
Family care burden
Parent			−0.00352	−0.0337 **	−0.0237 *
			(0.00771)	(0.0149)	(0.0141)
Number of children (under 18)				0.0172 ***	0.0141 **
				(0.00659)	(0.00619)
Financial literacy
FL score					−0.0338 ***
					(0.00123)
Country dummy	no	yes	yes	yes	yes
Observations	18,821	18,821	15,568	9434	9434
Wald chi2	894.903	3333.51	2697.97	1853.6	2316.89
p > Chi2	0.000	0.000	0.000	0.000	0.000
Correctly classified	0.879	0.875	0.8524	0.8188	0.8329

Standard errors in parentheses: *** p < 0.01, ** p < 0.05, * p < 0.1. In the table, we propose the estimates with different controls in model A we consider some personal characteristics of the respondents in B we add controls for countries, in C we include more controls for families’ responsibilities and characteristics of the place of residence, in D we also consider the number of children under 18 years old in the household, while in E we add the level of financial literacy of the respondents.

reports the marginal effects of the probit model run on the entire population. All the variables included in the model are significant and respect the initial hypothesis in terms of the signs assessing the positive or negative contribution of each of them. The likelihood ratio test statistic results of the model indicate that most of the variables are statistically significant at the 1% level of probability (p < 0.01).

The result of the model allows for the assessment of hypothesis (a), showing that being a woman has a significant positive effect on the probability of being a target of economic violence (coefficients ranging from 0.0204 to 0.0316 across different specifications, all significant at the 1% level). By adding explanatory variables to our model (see columns A through E in Table 8), the fact of being a woman remains significant and of negative sign. This leads us to be able to say that our hypothesis (a) is confirmed, i.e., that women are at higher risk of economic violence.

Moreover, as suggested by the literature (see paragraph 2), both age and education level—related to the ability to process more complex information and make decisions—show a negative relationship with the probability of being at risk. Specifically, younger individuals (aged 18–29, 30–39, 40–49, and 50–59) are significantly more likely to be targets (with coefficients ranging from −0.0914 to −0.0152, all significant at various levels), while older individuals tend to avoid being targets.

Education shows a protective effect, with higher levels of education associated with a lower probability of economic violence: primary school (−0.0309 to −0.162), lower secondary school (−0.0430 to −0.231), upper secondary school (−0.0905 to −0.299), university (−0.162 to −0.406), and post-graduate (−0.206 to −0.456), all significant at the 1% level for most categories.

The urban context also impacts the likelihood of being at risk. Individuals from larger cities are significantly less likely to be at risk compared to those from smaller towns (coefficients for various urban categories range from −0.0250 to −0.0463, all significant at various levels).

Conversely, being identified as a migrant indicates that you are potentially more at risk of economic violence. The coefficients for migrants are positive (ranging from 0.0221 to 0.0380), though the significance varies, indicating that migrants are more likely to be financially fragile and vulnerable.

Parenthood status alone does not show a statistically significant relationship with economic violence risk. However, the number of children to care for is significant and positively related to the risk (coefficients 0.0141 to 0.0172, significant at the 5% and 1% levels, respectively). This indicates that having more underaged children increases financial fragility, thereby increasing the probability of becoming a victim of economic and financial violence.

To test hypothesis (b), assessing gender gaps in the impact of the determinants of the risk of economic violence, we ran the probit model separately for women and men (

Table 9. Determinants of the probability of being at risk of economic/financial violence, women.

	A	B	C	D	E
Personal characteristics
Age, 18–29 years	−0.0227	−0.0567 ***	−0.0436 *	−0.039	−0.0125
	(0.0171)	(0.0200)	(0.0244)	(0.0319)	(0.0283)
Age, 30–39 years	−0.0599 ***	−0.102 ***	−0.0845 ***	−0.0727 **	−0.0347
	(0.0167)	(0.0194)	(0.0238)	(0.0313)	(0.0279)
Age, 40–49 years	−0.0600 ***	−0.0962 ***	−0.0833 ***	−0.0628 **	−0.0212
	(0.0166)	(0.0193)	(0.0238)	(0.0314)	(0.0280)
Age, 50–59 years	−0.0450 ***	−0.0916 ***	−0.0725 ***	−0.0499	−0.00908
	(0.0168)	(0.0194)	(0.0239)	(0.0314)	(0.0281)
Age, 60–69 years	−0.0179	−0.0792 ***	−0.0549 **	−0.0263	0.00462
	(0.0173)	(0.0196)	(0.0243)	(0.0321)	(0.0287)
Age, 70–79 years	0.0151	−0.0600 ***	−0.0214	0.0233	0.0389
	(0.0192)	(0.0206)	(0.0260)	(0.0348)	(0.0312)
Education level, primary school	−0.0704 **	−0.172 ***	−0.163 ***	−0.211 ***	−0.114 **
	(0.0329)	(0.0381)	(0.0484)	(0.0591)	(0.0574)
Education level, lower secondary school	−0.134 ***	−0.208 ***	−0.194 ***	−0.208 ***	−0.0795
	(0.0320)	(0.0377)	(0.0482)	(0.0586)	(0.0574)
Education level, upper secondary or high school	−0.139 ***	−0.283 ***	−0.277 ***	−0.309 ***	−0.144 **
	(0.0317)	(0.0373)	(0.0475)	(0.0576)	(0.0567)
Education level, university	−0.196 ***	−0.353 ***	−0.363 ***	−0.413 ***	−0.218 ***
	(0.0319)	(0.0374)	(0.0477)	(0.0580)	(0.0575)
Education, post-graduate	−0.236 ***	−0.378 ***	−0.386 ***	−0.450 ***	−0.249 ***
	(0.0321)	(0.0380)	(0.0486)	(0.0592)	(0.0590)
Migrant	0.0393 **	0.0318 *	0.0338 *	0.0347	0.0278
	(0.0170)	(0.0169)	(0.0200)	(0.0239)	(0.0226)
Urban, small-town			−0.0244 **	−0.0113	−0.0198
			(0.0113)	(0.0165)	(0.0156)
Urban, town			−0.0164	−0.0205	−0.0164
			(0.0105)	(0.0135)	(0.0128)
Urban, city			−0.0364 ***	−0.0481 ***	−0.0385 ***
			(0.00998)	(0.0126)	(0.0121)
Urban, large city			−0.00989	−0.0204	−0.00968
			(0.0148)	(0.0198)	(0.0190)
Family care burden
Parent			0.00441	−0.017	−0.0122
			(0.00995)	(0.0152)	(0.0144)
Number of children (under 18)				0.0142 **	0.0123 **
				(0.00562)	(0.00528)
Married and inactive	0.127 ***	0.122 ***	0.143 ***	0.185 ***	0.174 ***
	(0.00740)	(0.00754)	(0.00913)	(0.0117)	(0.0112)
Financial literacy
FL score					−0.0303 ***
					(0.00144)
Country dummy	no	yes	yes	yes	yes
Observations	15,386	14,273	10,192	6772	6772
Wald chi2	734.86	2064.85	1673.29	1218.1	1446.85
p > Chi2	0.000	0.000	0.000	0.000	0.000
Correctly classified	0.8584	0.864	0.8431	0.8153	0.8286

Standard errors in parentheses: *** p < 0.01, ** p < 0.05, * p < 0.1. In the table, we propose the estimates with different controls in model A we consider some personal characteristics of the respondents in B we add controls for countries, in C we include more controls for families’ responsibilities and characteristics of the place of residence, in D we also consider the number of children under 18 years old in the household, while in E we add the level of financial literacy of the respondents.

for the women population and

Table 10. Determinants of the probability of being at risk of economic/financial violence, men.

	A	B	C	D	E
Personal characteristics
Age, 18–29 years	−0.115 ***	−0.112 ***	−0.135 ***	−0.109 ***	−0.0792 **
	(0.025)	(0.0252)	(0.0291)	(0.0407)	(0.0342)
Age, 30–39 years	−0.139 ***	−0.141 ***	−0.158 ***	−0.108 ***	−0.0585 *
	(0.0248)	(0.0248)	(0.0289)	(0.0407)	(0.0344)
Age, 40–49 years	−0.144 ***	−0.145 ***	−0.162 ***	−0.117 ***	−0.0650 *
	(0.0246)	(0.0246)	(0.0287)	(0.0402)	(0.0341)
Age, 50–59 years	−0.124 ***	−0.128 ***	−0.147 ***	−0.0970 **	−0.037
	(0.0249)	(0.0248)	(0.0288)	(0.0404)	(0.0344)
Age, 60–69 years	−0.0890 ***	−0.113 ***	−0.128 ***	−0.0591	−0.00763
	(0.0256)	(0.0251)	(0.0291)	(0.0413)	(0.0352)
Age, 70–79 years	−0.0844 ***	−0.109 ***	−0.122 ***	−0.0524	−0.0218
	(0.0272)	(0.0261)	(0.0306)	(0.0437)	(0.0373)
Education level, primary school	−0.159 **	−0.133 **	−0.154 **	−0.158 *	−0.00253
	(0.0667)	(0.0581)	(0.0703)	(0.0863)	−0.0724
Education level, lower secondary school	−0.250 ***	−0.217 ***	−0.245 ***	−0.268 ***	−0.0811
	(0.0654)	(0.0571)	(0.0695)	(0.0654)	−0.0571
Education level, upper secondary or high school	−0.250 ***	−0.278 ***	−0.318 ***	−0.298 ***	−0.170 **
	(0.0653)	(0.0565)	(0.0686)	(0.0655)	−0.057
Education level, university	−0.298 ***	−0.332 ***	−0.377 ***	−0.451 ***	−0.170 **
	(0.0653)	(0.0565)	(0.0686)	(0.0818)	(0.0701)
Education, post-graduate	−0.323 ***	−0.357 ***	−0.404 ***	−0.501 ***	−0.216 ***
	(0.0661)	(0.0570)	(0.0692)	(0.0824)	(0.0718)
Migrant	0.0180	0.0136	0.0235	0.0358	0.0285
	(0.0216)	(0.0197)	(0.0222)	(0.0310)	(0.0296)
Urban, small-town			−0.0317 **	−0.0431 **	−0.0444 **
			(0.0129)	(0.0218)	(0.0199)
Urban, town			−0.0343 ***	−0.0463 ***	−0.0335 **
			(0.0115)	(0.0173)	(0.0163)
Urban, city			−0.0294 ***	−0.0321 *	−0.0252
			(0.0112)	(0.0165)	(0.0154)
Urban, large city			−0.0386 **	−0.0800 ***	−0.0546 **
			(0.0162)	(0.0235)	(0.0239)
Family care burden
Parent			−0.0158	−0.0526 **	−0.0401 *
			(0.0108)	(0.0243)	(0.0230)
Number of children (under 18)				0.0163	0.0147
				(0.0108)	(0.0102)
Married and inactive	0.213 ***	0.170 ***	0.214 ***	0.283 ***	0.230 ***
	(0.0235)	(0.0215)	(0.0293)	(0.0453)	(0.0410)
Financial literacy
FL score					−0.0350 ***
					(0.00181)
Country dummy	no	yes	yes	yes	yes
Observations	8812	8812	7255	3939	3939
Wald chi2	454.03	1402.13	1076.72	649.16	905.88
p > Chi2	0.000	0.000	0.000	0.000	0.000
Correctly classified	0.9082	0.8957	0.837	0.8705	0.8502

Standard errors in parentheses: *** p < 0.01, ** p < 0.05, * p < 0.1. In the table, we propose the estimates with different controls in model A we consider some personal characteristics of the respondents in B we add controls for countries, in C we include more controls for families’ responsibilities and characteristics of the place of residence, in D we also consider the number of children under 18 years old in the household, while in E we add the level of financial literacy of the respondents.

for the men population). In this second step, a more specific variable attesting to whether a subject is both married and outside the labor force was added. For this variable, the hypothesis is that being in this fragile position is a key determinant of being at risk of economic violence.

Table 9 and Table 10 illustrate gender differences in exposure to the risk of economic violence, which can also lead us to validate our hypothesis (b), namely that there are gender differences in the determinants of economic violence risk. In particular, our analysis asserts that for women, being a migrant is significant and positively related to the risk (coefficients ranging from 0.0278 to 0.0393, significant at various levels), while it is not significant for men. The family care burden, identified by the number of children under 18, has a significant and positive effect for women (coefficients 0.0123 to 0.0142, significant at the 5% level), but is not significant for men.

Moreover, the protective effect of education is systematically greater for women than for men. For example, the coefficients for university education range from −0.196 to −0.413 for women, all significant at the 1% level, compared to men, where the effect, though present, is less pronounced.

Finally, regarding our last hypothesis to be tested, we can confirm that the impact of financial literacy is positive and significant; consequently, we can validate our hypothesis (c) by stating that financial literacy has a protective effect against the risk of being a target of economic violence (see column E of Table 8). Also interestingly, this positive effect is the same for both women and men (see column E of Table 9 for women and column E of Table 10 for men). The general positive effect of financial literacy in preventing economic violence is strong and equally applies to both genders (financial literacy coefficients for women −0.0303, significant at the 1% level, and for men −0.0350, significant at the 1% level). This suggests that financial literacy’s preventive effect is robust across genders, even though barriers to acquiring financial literacy are different.

The results of this study shed light on the intricate relationship between gender disparities, educational attainment, and labor market outcomes. The analysis demonstrates a notable correlation between higher educational attainment and improved labor market outcomes, though gender disparities remain evident.

Our findings confirm that women are at a higher risk of economic violence compared to men, consistent with our first hypothesis. This aligns with previous research indicating that women’s financial vulnerability is influenced by socio-economic and cultural factors that disproportionately affect them. The probit model reveals a significant positive coefficient for the variable ‘Women’, suggesting that gender remains a robust determinant of economic violence even after controlling for other factors.

The results indicate that age and education level play crucial roles in mitigating the risk of economic violence. Specifically, older individuals and those with higher levels of education are less likely to be at risk. This suggests that both age and education contribute to greater financial autonomy and awareness, which can shield individuals from economic exploitation. For instance, our data shows that younger individuals, particularly those aged 18–29, have a higher probability of being targeted, which may reflect their relative financial inexperience and potential for economic dependency.

Migration status emerges as a significant factor, with migrants facing higher risks of economic violence. This finding aligns with previous studies that have highlighted the vulnerabilities faced by migrants, including job insecurity and limited access to financial services. Our results suggest that policy measures aimed at improving financial stability and literacy among migrant populations could be crucial in reducing their risk of economic violence.

The probit model reveals several important determinants of economic violence. Education consistently shows a protective effect, with higher levels of education being associated with reduced risk. This underscores the value of educational interventions in preventing economic violence. Similarly, living in larger urban areas is linked to lower risk, which might be attributed to better access to financial resources and support services in cities.

Our findings indicate that while both men and women benefit from higher education in terms of employment rates and earnings, the advantages are more pronounced for men. This suggests that despite achieving similar educational levels, women may still encounter barriers in the labor market. These barriers could include discrimination, societal expectations, and unequal access to networking opportunities.

For instance, the data shows that men with a bachelor’s degree tend to have higher median earnings compared to women with the same level of education. This wage gap persists across different educational levels, emphasizing the need for policies that address wage equality and support women’s career advancement.

The analysis also highlights a positive relationship between higher educational attainment and employment rates for both genders. However, the employment rate for women, even with higher education, is lower than that for men. This could be attributed to factors such as caregiving responsibilities, part-time work preferences, or challenges in balancing work and family life. Policies that provide better childcare support and promote flexible working arrangements could help mitigate these issues.

While higher educational attainment is associated with better labor market outcomes for both genders, significant gender disparities persist. Addressing these disparities requires comprehensive policies and practices that promote gender equality in education and employment. By implementing strategies to support women’s career advancement and reduce the wage gap, we can create a more equitable and inclusive labor market. Thus, the results of this study highlight the critical need for continued efforts to achieve gender parity in both education and the workforce.

The positive impact of financial literacy on reducing the risk of economic violence supports our third hypothesis. Our analysis demonstrates that higher financial literacy is associated with a lower probability of experiencing economic violence, regardless of gender. This reinforces the idea that financial knowledge and skills are crucial for financial resilience and protection against economic abuse. The uniform effect of financial literacy across genders highlights its universal importance, although it is noteworthy that the barriers to acquiring financial literacy may differ between men and women. Our results highlight the protective role of education and financial literacy. Educational programs to enhance financial literacy should be prioritized, particularly for younger individuals and those in precarious financial situations. By equipping individuals with financial knowledge and skills, we can improve their ability to manage financial resources effectively and reduce their exposure to economic violence.

5. Conclusions

The article offers an initial international quantification of the risk of economic violence (REV). Through this measure of economic violence risk, it was possible to verify with real data the fact that women are at greater risk of economic violence, even though the exposure differs depending on the country considered. The higher risk percentages among women in almost all the countries studied suggest that financial dependence and restricted access to economic resources are still pervasive issues. In addition, the impact of factors that may protect against greater vulnerability is different for men and women: the impact of education has a greater protective effect for women than for men; furthermore, for women, factors of greater exposure to the risk of experiencing economic violence are related to unpaid care work within households (i.e., the presence of children under 18 cohabiting in the household, having a partner, and not being active in paid work) and migrant status. Further exploration into gender differences reveals that young women with lower educational attainment, migrant status, residing in smaller urban areas, and shouldering caregiving responsibilities are at heightened risk of economic and financial violence. Particularly striking is the correlation with low levels of financial literacy, which significantly amplifies this risk.

Given the gender differences in the phenomenon, it is, therefore, necessary to take proactive measures to effectively prevent and combat economic violence and to monitor its prevalence and differences at the international level by adopting, implementing, and monitoring primary and secondary prevention measures aimed at addressing gender inequality as a root cause of economic violence. To this end, it is essential to improve data collection and address economic violence more broadly in the international context. Conducting regular demographic surveys on economic violence and behaviors that increase the risk of being a target of economic violence could allow for a better definition of prevalence, causes, and consequences through an intersectional approach. In addition, funding for measures to prevent and address economic violence at the local, national, and international levels should be an integral part of activities to reduce violence against women, as should the introduction of legislative measures to criminalize economic violence, hold perpetrators accountable, and protect victims. It is necessary to emphasize the need for a global perspective on the phenomenon, which requires international collaborations or frameworks that could help standardize measures to combat economic violence in different countries, as well as the establishment of long-term strategies that could help mitigate economic violence, such as promoting gender equality in education and employment, improving access to financial services, and enhancing economic opportunities for women.

Finally, having found the positive impact of financial literacy in preventing exposure to the risk of experiencing economic violence, we believe that investment in financial education campaigns targeting women, especially in countries at higher risk of economic violence, is essential. Furthermore, the geographical variations in risk levels point to the need for context-specific policy interventions. In countries with high levels of economic violence, there is a pressing need for policies that address the structural barriers to financial independence for women. This could include initiatives to improve access to education and employment opportunities, enhance social protection systems, and promote gender equality in all spheres of life. Finally, by addressing gender disparities in financial literacy and empowering women with the necessary knowledge and skills, policymakers can advance more inclusive strategies aimed at reducing economic violence.

Within the limitations of the available data and the analysis conducted, our article proposes a first attempt at an international analysis of the phenomenon of economic violence. Our quantification of the risk of economic violence also provides an innovative element to be integrated into attempts to assess and quantify individuals’ levels of well-being and financial resilience. In fact, to date, these measures often only consider subjective perceptions and, more generally, do not take into account the heterogeneity of individuals in terms of their ability and freedom to make decisions on specific economic and financial issues. Consequently, the measure of the risk of economic violence proposed in this article can be integrated into the conceptualization and quantification of a truly inclusive measure of financial well-being.

However, many questions remain open for new research horizons, for example, trying to expand the geographic coverage of the analysis of economic violence, which in our analysis does not adequately cover Africa and Asia, to provide a more global perspective on the phenomenon. Another interesting avenue to pursue is the intersectional analysis of the long-term impact of economic violence by determining its effects on different population groups or still defining which types of financial literacy programs are most effective, considering the gender disparities highlighted in our study.