Shared Physical Custody and Child Maintenance Arrangements: A Comparative Analysis of 13 Countries Using a Model Family Approach

This book chapter provides new insights to the question of how child maintenance policies have responded to changing post separation family arrangements and most specifically shared physical custody (SPC). We analyse how SPC is implemented and how it operates in child maintenance policies in 13 countries: Australia, Belgium, Denmark, Estonia, Finland, France, Iceland, New Zealand, Norway, Spain, Sweden, the UK and the U.S. The comparative analysis is based on vignette questionnaire collected in 2017. There are differences in how countries have acknowledged and recognized shared physical custody in their child maintenance policies. It varies from complete annulment of obligations, to some countries making finer grained adjustments to reduce child maintenance obligations and yet others’ making no changes as a result of shared physical custody, with the paying parent still having to provide the full amount of child maintenance. It seems there is no standard practice and nor do the different arrangements map easily onto child maintenance scheme typology. The latter is surprising, as it might have been expected that similarly structured child maintenance schemes would treat shared physical custody in similar ways. This variability demonstrates a lack of coherence across child maintenance policies on how to deal with this phenomenon of greater gender equality in post-separation parenting arrangements.


Introduction
Most western countries have a variety of policies designed to secure incomes after parental separation, including setting child maintenance obligations, which is the financial contribution to be paid by a non-resident parent to a resident parent for supporting children post separation (International Network of Child Support Scholars 2019). In the vast majority of cases worldwide, and irrespective of welfare regime, these payments are made from a minority care-time non-resident father to a majority care-time, resident mother. In some countries, if parents are unable or unwilling to pay, the state may provide guaranteed or advanced maintenance (Corden 1999;Skinner et al. 2007Skinner et al. , 2012. In the face of rising rates of family breakdown, through divorce or separation, more families become subject to child maintenance policies, making this an increasingly important policy aspect of modern family life. Also, as we have seen throughout this book, shared physical custody has become more common in separated families. Certainly, a growing number of separated parents jointly share the care of their child(ren) either equally, or at least 30% of care by each parent (Fehlberg et al. 2011;Trinder 2010;Smyth 2017;Hakovirta and Eydal 2020). Multiple terms are used for this phenomenon, including shared care, shared residence or joint physical custody. 1 For the purposes of standardisation we use shared physical custody (SPC) throughout this chapter. It means that the child spends equal time living with both parents and both parents physically care for the child. Shared physical custody arrangements however, also signify a greater ambiguity in family roles and responsibilities as well as more fluidity in living arrangements as children live with both their parents separately and move across their parents' households (Cancian et al. 2014;Carlson and Meyer 2014).
In turn, this creates more family complexity presenting substantial operating challenges for child maintenance policies; that is if they pay any regard to care arrangements at all. Certainly, the traditional breadwinning father is no longer the norm in many countries, because more mothers stay in employment post childbirth and dual earner families are more common. Child maintenance policies are of interest because they have to deal with this fluidity and complexity in care arrangements and this has direct consequences for the economic well-being of children and their parents. Such policies also embody a set of values regarding parental responsibility post-separation (Skinner et al. 2007). Yet, we know little about how countries deal with this challenge of accounting for the sharing of care between parents, especially what happens when that share is nearly equal. Nor do we know about the policy principles that might guide operations, or the policy adaptions, or how they might differ across countries, or what these differences might signify.
This chapter intends to fill some of that gap in knowledge. It will provide answers to the questions of how child maintenance policies deal with the sharing of responsibility between separated parents and whether the guidelines in child maintenance policies take account of the degree to which the other parent engages in care. We take a comparative approach across 13 countries in total: Australia, Belgium, Denmark, Estonia, Finland, France, Iceland, New Zealand, Norway, Spain, Sweden, the UK and the U.S. 2 These countries represent different child maintenance schemes (Skinner and Davidson 2009) which we describe in more detail in the next section.
The analysis here updates and extends the work of Skinner et al. (2007) who produced one of the first comparative studies that considered shared physical custody arrangements in child maintenance policies across 14 countries, albeit briefly. Also, it extends the work of Skinner et al. 2012, who looked more indepth across five countries (Finland, Iceland, Netherlands, UK and U.S). The study reported here adapts Skinner et al.'s (2007) original model families approach to look in more detail at shared physical custody arrangements and child maintenance rules and formulae as well as across a different set of countries (which now includes Spain, Estonia and Iceland). In the model family method used in this 2017 study, national experts are presented with fictitious families in a range of different circumstances and are asked to provide information on policy responses relating to those circumstances. This exposes how policy can operate through the application of decisions, rules and guidelines in a set of proscribed circumstances to produce different outcomes (in this case child maintenance amounts). By adapting this method we work out what the child maintenance amounts might be for different shared physical custody arrangements and therefore the research reported here extends the documentary policy analysis of child maintenance schemes provided by Claessens and Mortelmans (2018) in eight countries (Canada, Denmark, Finland, France, Germany, Netherlands, Sweden and UK). This chapter will add to this emerging body of comparative evidence by providing new insights into how countries deal with (or fail to deal with) family complexity as society and policies adapt to new post-separation parenting arrangements. Too little is known about the intersection of these two issues despite; a growing policy interest in some countries, a large amount of research exploring the changing nature of family and gender relationships and a growing number of studies on child maintenance policies.

Typologies of Child Maintenance Schemes
The reorganisation of parental relations after separation or divorce assumes the right of the child to receive maintenance from both parents. From the parental point of view, it is the legal responsibility of both parents to take charge of the child's care, education and maintenance in accordance with their abilities, regardless of whether the child is in the care of one or both parents (Wikeley 2009). Parental obligations after separation or divorce differ substantially between countries in terms of their underlying philosophy, structures, rules and organisation and in particular produce very different outcomes.
Notably, there have been a few attempts to provide comparative analysis of child maintenance schemes including the early pioneering work of Millar and Warman (1996) and Corden (1999). Millar and Warman studied family obligations across nine European countries and explored whether there were common trends towards new definitions of family obligations in the context of changing family structures and relationships. They did not construct a typology but their main conclusion was that agreements about financial arrangements post separation relied mainly on private agreements between parents and in some countries these were ratified by the courts. Some countries relied on standard rules or guidelines while in others, cases were dealt with individually on a discretionary basis. Six of the nine countries they compared had some form of guaranteed maintenance scheme. Corden (1999) compared child maintenance regimes in ten European countries: Austria, Belgium, Denmark, Finland, France, Germany, the Netherlands, Norway, Sweden and the UK. She found each regime developed from a different legal and historical background, but the general pattern was towards equal treatment for all children in respect of child maintenance, irrespective of the marital status of their parents. Each country had different structural and administrative arrangements and decisions about whether to pay and how much child maintenance to pay, were made variously by parents themselves (with or without help), by court judges or officials, or by administrative staff in social security or welfare offices. The UK and the Netherlands at that time were the only countries which had no specific scheme to advance child maintenance, apart from general social assistance benefits.
As previously mentioned, Skinner et al. (2007) undertook a large cross-national analysis of child maintenance schemes in 14 countries. They considered the logics of formal decision making, the determination of child maintenance obligations and the enforcement and penalty provisions used in the event of non-compliance. They clustered countries according to the weight given to the court and/or agency in setting child maintenance orders using the data they collected in 2006. Three maintenance schemes emerged and were identified as operating a court, agency or hybrid scheme. In Austria, Belgium, Canada, France, Germany and Sweden, courts had the main responsibility for the determination of formal child maintenance obligations. In Australia, Denmark, New Zealand, Norway and the UK, an administrative agency was responsible for assessment, collection and transfer of child maintenance payments. These countries represent the agency model. In Finland, the Netherlands and the U.S., responsibility for the determination of child maintenance obligations lay with several institutions, for example with the municipal welfare board and/or the court. Generally, Skinner et al. (2007) found that court based schemes operated on a more discretionary basis and cases were treated individually, whereas agency and hybrid schemes tended to take more standardised approaches and applied formulae and rules in the decision-making process.
For the purposes of comparison, we clustered our countries according to the typology developed by Skinner et al. (2007) based on the different institutional loci of decision-making. Thus, Australia, Denmark, New Zealand, Norway and the UK were categorized as agency schemes, the same as before. Belgium, Estonia, Spain, Sweden and France were categorized as court based schemes, as courts had the main responsibility for the determination of formal child maintenance obligations. Finland, Iceland and the U.S., were considered as hybrid schemes because they locate their decisions regarding child maintenance in several institutions which are an amalgamation of courts and agencies.
We used this typology as it is reasonable to expect that the different institutional settings might have a bearing on how the sharing of care time is accounted for in calculating child maintenance obligations. Whilst our research study is exploratory and descriptive, it is possible to consider that court based systems are more discretionary and are more likely to recognise shared physical custody arrangements as they tend to treat cases on an individual basis and in that regard will follow changing social norms. Whereas, agencies generally apply more fixed rules and formulae and may be less likely to respond to changing social norms and rising trends in shared physical custody arrangements, because to do so may require legislative changes to operational procedures and this could inhibit responsiveness and adaptation. However, where agency schemes do recognise shared physical custody, it might be they take a more standardised approach producing similar outcomes across countries compared to court based schemes.

Prior Research on Shared Physical Custody and Child Maintenance
In many countries a major legal premise in family law is that children should share time with both parents after separation (CRC 1989). However, sharing care of children, beyond traditional gendered and more limited visitation arrangements, is more complex than present guidelines in child maintenance policies recognize (Melli and Brown 1994;Beld and Biernat 2002;Bartfeld 2011;Claessens and Mortelmans 2018). Claessens and Mortelmans' (2018) documentary analysis of eight countries revealed that the shared physical custody arrangements are accounted for in child maintenance policies in various ways, some of which can be highly accommodating and others disadvantageous for the modern post-divorce family. They suggested that policy concerning gender equality in shared physical custody arrangements does not consistently translate into child maintenance policies. In the U.S. almost all states explicitly address shared physical custody in their child maintenance guidelines and typically produce orders that are lower than would be the case under other time sharing arrangements (Brown and Brito 2007). Other research in the child maintenance context suggests that shared physical custody does not necessarily lead to fathers providing financial support for their children (i.e. in the form of child maintenance to the other parent) and the obligation can be annulled in some cases (Singer 2008;Hakovirta and Rantalaiho 2011). Yet, qualitative evidence suggests that mothers often carry more of the responsibility than their former partners for management of children's daily lives, including paying school-related expenses, medical, and dental costs (Cashmore et al. 2010;Lacroix 2006). In Australia Lodge and Alexander (2010) found that everyday expenses were usually paid by the parent they lived with for most of the time. In the case of equal time parenting, the 'vast majority' of adolescents said that both parents made a contribution to their everyday expenses.
Less is known however, about how child maintenance policy works in practice within and across countries and what the levels of child maintenance payments are if children are in shared physical custody arrangements. Skinner et al. (2007) compared maintenance awards under shared physical custody for two children in £ppp 3 per month in 2006. They reported that the highest maintenance award in those countries where maintenance was expected to be paid was in Canada and the U.S. In Australia, France, Norway, New Zealand and the UK obligations were lower, but the parent who had higher income still paid maintenance. In Belgium, Denmark Finland, Netherlands and Sweden the child maintenance obligations were annulled. The other study by Skinner et al. (2012) compared the effect of shared care on child maintenance amounts. In comparison to the situation where children had two weekend visits, in shared physical custody situations, the amount was substantially reduced in the U.S., only reduced a little in Finland, whereas in the UK the obligation to pay was eliminated completely. In Iceland shared physical custody had no effect on maintenance awards. Proponents that argue for a shared physical custody presumption being embedded in family law and family policies hope it will benefit children by promoting both parents' continued involvement and encouraging them to share more equal responsibility for raising their children. The evidence to date shows how more equal care arrangements may produce different financial impacts and we aim to investigate this further with a deeper systematic comparative analysis with new research data collected in 2017.

Methods
The aim of this section is to explore whether shared physical custody arrangements are taken into account in child maintenance policies and if so, in what ways does this happen and how does it vary across countries?
We use a model family approach in which national informants complete a detailed standardised questionnaire providing information on their policy. They were asked to describe their child maintenance policy and to calculate the amount of child maintenance in the prescribed hypothetical model families according to their own country's policies and legal guidelines. This method has been used successfully to make comparisons of the tax/benefit package for families (e.g. Bradshaw and Finch 2002;see Bradshaw 2009 for an overview). As part of the model families approach we generated a number of vignettes that are short stories of fictitious families that provide fixed details of family situations (see Soydan and Stål 1994;Barter and Renold 1999). Vignettes are the component part of the model family approach and have been used successfully in many comparative studies on child maintenance policies (e.g. Corden 1999;Skinner et al. 2007Skinner et al. , 2012Skinner et al. , 2017Meyer et al. 2011;Meyer and Skinner 2016;Hakovirta and Eydal 2020). Vignettes represent real-life situations in meaningful social circumstances, and the national informants (who are our respondents) could then offer their observations and interpretations from within their own policy contexts, but for standardised family types. This ensures that as far as possible, like is being compared with like and standardized comparison across countries is reliable as the stimuli is held constant for national informants across the countries.
Data was collected at the end of 2017. We recruited national informants through professional contacts in the research community. Mostly there was one informant from each country. Many of the national informants were academics who had earlier experience in similar studies either as informants, or were involved in collecting or analysing vignette data in previous child maintenance studies or based on their earlier contribution to the field. As each informant was an expert in the field of enquiry in their own country, it eased the task of data collection and validation and thereby helped provide a deeper and insider's interpretation of the policy framework and the operational rules and processes.
National informants completed a detailed standardised questionnaire providing information on shared physical custody and child maintenance policy. They were also asked to calculate the amount of child maintenance the law required the parent to pay in the prescribed model families according to their own county's policies and legal guidelines. Note that the calculated child maintenance amounts they produced were related to the model family and their current circumstances, which were fixed at a certain point in time.
Vignette method has some limitations. First limitation, which is typical to this type of research, is that the data are from only one policy expert in each country; including multiple experts within a country could lead to more confidence in the policy descriptions. Second, we do not have information from court experts who deal with the issues in child maintenance cases. Finally, we focus here only on the level of obligation, and this amount may or may not be paid. The data therefore highlight how policy works in these particular model family situations.
In the vignette we first provided a basic situation (Base Case 'A') which included information for national informants to describe how their policy works and all necessary information to calculate the child maintenance obligation. Our vignette story was as follows: Mary and Paul are getting a divorce after ten years of marriage. They have two children. Emily, aged seven, and Sophia, aged ten. Both children attend a local school and there are no school fees. After the divorce, Mary and the children will continue to live in the rented apartment Mary and Paul shared during marriage so that the children can stay in their home. Paul will rent a new apartment in the same suburb nearby. The number of bedrooms, rent and other housing costs of both apartments are averages that are typical in your country. Paul is employed and earns median monthly male full-time earnings for your country. Mary is also working, earning median monthly female full-time earnings for your country. Mary and Paul have agreed that they will have 'joint legal custody' of the children, sharing the major decisions affecting the children. In terms of living arrangements, Emily and Sophia will have two overnight stays from Friday afternoon until Sunday afternoon at their father's home every other weekend.
In the base case both parents were working full time and had median incomes that were typical (median monthly earnings) in their country. In that way the parents are presented as being on a level playing field in that both are earning typical wages for full-time employees, except of course commonly there is a gender pay gap which will vary across countries. These gender inequalities are automatically reproduced here as we use gender specific median income amounts in the vignettes. We then asked the informants to calculate the outcome in terms of whether there would be a formal child maintenance arrangement, and if so, the monthly amount that would be awarded in these circumstances. In the next scenario of the same vignette, the situation is otherwise exactly the same as in base case 'A', but Mary and Paul had a shared physical custody arrangement in which the children spend exactly an equal amount of time with both parents. Every other week is spent with Mary and every other week with Paul. We asked the experts to explain how the outcomes would differ now that the parents had an equal shared physical custody arrangement. This equal care scenario would represent an ideal of equality with a presumption of 50:50 care-time and one that is perfectly and consistently exercised by parents. In that regard model family approaches cannot take account of the messy reality of families' lives in which arrangements may vary frequently. That is both the strength of the model family approach (standardisation) and its potential weakness as it can only give an approximation of reality.
We conducted the analysis in three main ways: First, we provide the country context and present reported prevalence rates and definitions of shared physical custody (Table 14.1). Second, we analyse the answers to the questions on child maintenance policy and shared physical custody exploring how it was acknowledged in child maintenance policy, highlighting variations in approaches (Table 14.2). Third, we are using the informants' calculations of the amounts of child maintenance liabilities when the children in the model family had two overnight stays with their non-resident parent every other weekend and compared that to when there was equal shared physical custody (keeping parental incomes the same as in the base case). We calculate the financial outcomes by modelling child maintenance amounts in pppUS$ across countries, facilitating a meaningful cross country comparison of policy outcomes (Fig. 14.1). The analysis is therefore Source: Column 1 and 2 are based on relevant national surveys or statistics as reported by national informants in each country (cited in text) and column 3 depends on information collected from the respondents numbers are not fully comparable as the information comes from different sources in each country (official statistics, administrative records or surveys). N/A not available

Prevalence of Shared Physical Custody
We start the analysis by presenting the 'care context' and the information provided by national informants on what they know about the prevalence of shared physical custody. Defining shared physical custody is difficult as the definition is broad and can be used to cover a range of care arrangements. Very often comparative work on shared physical custody arrangements is bedevilled by different terms, definitions, time thresholds, measures, and units of analysis which means that cross-national comparisons and research translation present formidable challenges. In general, it refers to a sharing of care time of children between parents, but the care-time can range from 25% to 50% spent with each parent (see Fehlberg et al. 2011;Smyth 2017;Trinder 2010). Also the source of information on prevalence matterswhether the information comes from official statistics, administrative records or surveys. For example, many studies are reliant on divorce records to estimate the incidence of shared physical custody arrangements. To some extent, these may underestimate the prevalence of shared physical custody as divorce records ignore the separations from The amount of child maintenance (per child) expected to be paid in a model family when a child stays with one parent two nights every other weekend versus the situation that parents have shared physical custody co-habitant relationships. In addition, some countries may have no readily available information.
Keeping these challenges in mind, Table 14.1 presents the prevalence rates of shared physical custody arrangements (column 1) and the various different time thresholds that these were based on (column 2) as reported by national informants. Informants' reports refer to a range of different sources (official records, surveys) and are therefore highly variable. Even so, to our knowledge they represent the most recent sources of information in each country and are the best available. On a cautionary note therefore, the variations in the ways shared physical custody is defined is important as it affects the prevalence rates presented. For example, prevalence may be higher if it is defined as each parent having care for at least 30% of the time rather than a 50/50 split.
So far only a few countries have adopted a legal presumption of 50/50 joint custody; for most countries there is no clear definition and it is only mentioned as an arrangement where children live an equal amount of time with both parents. It is better to think of the numbers in Table 14.1 therefore, as descriptive information and not data per se, as the latter term implies some sort of standardisation, which clearly it is not. We give more detail on sources from each country when reporting the analysis of the table. In relation to the last column 3 of Table 14.1, we also show the time thresholds used for shared physical custody, but taken from within each country's child maintenance scheme. Again, this is as reported by national informants, based on their knowledge of administrative rules and how legal institutions and judicial decision-making might work in practice. For the purposes of comparison, we have grouped countries in Table 14.1 by the type of maintenance scheme to see if there appears to be any common pattern between the reported prevalence rates (column 1) and official time thresholds used in different child maintenance schemes (column 3).
The results in Table 14.1 show that the reported rates vary markedly and there is no easily discernible pattern; which is not surprising given the range of information sources (Estonia had no data). However, what is of interest is that shared physical custody remains a minority arrangement for real separated families across all countries, with only one region of Spain having a high reported rate of 40%. Next we report the prevalence rates according to the child maintenance regimes even though the rates do not easily follow the maintenance typology.
In the agency regime with relatively lower prevalence rates are the UK and New Zealand. In the UK, reports from a number of different survey sources show that prevalence ranges from 3% to 17% depending on the source. Notably however, some accounts from resident parents suggest that 50-50 time arrangements could be as low as 1% (resource quoted as Haux et al. 2017). In New Zealand no robust information is available. Under the pre-2013 Child Support formula approximately 5% of cases were shared physical custody cases (i.e. each carer had at least 40% of care-time) but this does not include private arrangement cases which are not part of the formal child support system. Five percent can be seen as a lower bound, but it is unlikely that the true figure is many times that because shared physical custody in private agreements between parents may be defined more loosely as more-or-less equal care.
In the other countries among agency regimes about 20-25% have shared care arrangements. In Australia 20% of children under 18 years of age had shared physical custody arrangements, as reported by the one parent survey in 2012 (Qu et al. 2014). In Denmark, shared physical custody arrangements are not registered in administrative data and therefore are only accessible in surveys. The incidence of shared physical custody arrangements are recorded by the age of children. The proportion of all children of divorced parents reported to have shared physical custody in 2013 was 22% of 3-years-olds, 40% of 11-years-olds and 32% of 15-years-olds (Ottosen et al. 2014). In Norway when mothers' and fathers' responses are considered together (i.e. where both agree the child has shared physical custody) it accounted for 25% of children of separated parents according to survey data in 2012 (Kitterød and Wiik 2017).
In court based regimes the countries had the highest rates among all countries studied, expect in France. In France, shared care is considered when children spend roughly an equal amount of time with each parent. In 2012, 16.9% of children of divorced parents and children born out of wedlock are reported as having shared physical custody arrangements. Percentages are based on a survey of the decisions issued by Family Justice Judges (resource quoted as Belmokhtar 2014). Spain could also be considered as belonging to this lower group, but simultaneously also to the higher group, depending on the region. In Spain, the care arrangements postseparation are based on parenting plans which parents must submit to court and include the commitments they make regarding the custody, care and education of their children. Therefore, there is no precise definition and no specific threshold used to define shared physical custody. The figures are based on the number of divorce orders judges consider shared physical custody. There are striking regional disparities with proportions ranging from higher than 40% in Catalonia as opposed 8% in Extremadura in 2015. The average however, is 24.6% of the total number of divorces involving children where care is shared by parents (Flaquer et al. 2017) and this would put Spain in the higher group. In Belgium studies suggest that shared physical custody (defined as spending between 33% and 66% of the time with each parent) has become more popular in recent decades. Specifically, less than 10% of the children whose parents separated between 1990 and 1995 were in shared physical custody. By 2006 or later, 37% of children were reported to be in the care of each parent for at least 33% of the time (Vanassche et al. 2017). In Sweden, the Supreme Court has stipulated that arrangements with a less-than-equal split must generally be regarded as contact unless there are special factors pointing in the opposite direction (Newnham 2010). Shared physical custody therefore occurs for 35% of children of separated parents in Sweden in 2012/13 (SCB 2014).
In hybrid regimes, only in the U.S., Wisconsin, shared care has become popular post separation living arrangement of children. In the U.S. there is no national data. The national informant notes that the most recent data on divorce comes from court records in Wisconsin. Meyer et al. (2017) report that in divorces in 2010, 35-50% had shared physical custody, the lower percentage referring to 50/50 timeshare and higher percentage to 25% time share. In Finland, the reported prevalence rate is approximately 15% of children have shared physical custody. This is based on records from parents who have confirmed the child's residence agreement with the Social Welfare Board as having a shared physical custody arrangement (Child custody and maintenance 2017). However, not all parents in Finland confirm their child residence agreements with the Social Welfare Board and the extent to which this data reflects actual arrangements remains unclear. In Iceland, according to survey data, among divorced parents, 24% of children lived in shared physical custody (defined as 50/50 time share) (Júlíusdóttir 2009).
Time thresholds used to calculate the prevalence of shared physical custody in surveys and court records (shown in column 2 of Table 14.1) is not always the same as that used for determining child maintenance obligations. Therefore, it is important to see how within child maintenance schemes the measures might differ for recognising shared physical custody from that used in surveys or administrative records, and we report that in Table 14.1 column 3.
We can see in Table 14.1 that in five countries an equal time threshold of 50% is reportedly used to determine shared physical custody for child maintenance purposes (Norway, UK, France, Sweden and Iceland). This matches well with the idea of a gender equal split of parental responsibilities post-separation. In another five counties however, a range of time is used to determine the threshold for shared physical custody with most having a lower, more generous level than a 50% time share. So Australia, Denmark, and Belgium use a lower bound of about a third of time (35%, 36% and 33% respectively), whereas in NZ and Finland it was a bit higher (48% and 43% respectively). The U.S. (Wisconsin) child maintenance system appears to give the most generous recognition, setting its threshold at 25% of care time. In some countries (Spain and Estonia) it is numerically impossible to define as there is no standard threshold, or no regulations setting care time in child maintenance schemes.
Surprisingly perhaps, there are no clear similarities in the thresholds used by the type of child maintenance institutional arrangement. For example, it cannot be said that court based systems (which are generally more discretionary) were more likely to operate a more generous lower level for recognising shared physical custody than agencies (which generally apply more fixed rules and formulae). Given the discretionary nature of court based systems, it would have been reasonable to assume they would be quicker at responding to changes in social norms (such as rising trends in shared physical custody arrangements) than would be the case for administrative type child maintenance systems and therefore more likely to set lower thresholds for recognising joint physical custody. However, there is no evidence of that here using this data and this methodology of national informants. Moreover, when comparing the prevalence rates reported in column one of Table 14.1 with the thresholds used in child maintenance schemes in column three there are also no obvious patterns. This is also interesting, because it might have been expected that in countries which report higher prevalence rates of shared physical custody, the child maintenance schemes would have operated more generous lower time thresholds in recognition of this trend, but there is no evidence here of that either. There does seem to be some relationship between column two (thresholds used in reports to identify shared physical custody) and column three (thresholds used in child maintenance schemes). In 6 of the 13 countries, they correspond directly (Australia, New Zealand, UK, Belgium, Sweden and Iceland). This might suggest that maintenance schemes may have referred to available reports to set thresholds, but we do not know if this is the case. Suffice to say for now; there seems no obvious relationship between reported prevalence rates of shared physical custody and the reported time thresholds used for child maintenance purposes.
In the next section we consider in more detail how child maintenance schemes operate in taking account of shared physical custody and what potential effect this might have, such as whether the amounts of child maintenance is reduced for the shared physical custody scenario.

Accounting for Shared Physical Custody in Child Maintenance Policies
First, we begin our detailed analysis of child maintenance schemes by considering whether they treat parents equally in terms of assessing both their incomes for the purposes of determining how much maintenance should be paid. In recent years, counting both parents incomes to assess child maintenance liabilities has grown in popularity, what is called an 'incomes shares' approach. The income shares approach is considered to be more flexible and therefore more equipped to accommodate changing family realities and are sometimes cited as better able to accommodate shifts in shared physical custody (Cancian and Costanzo 2019). In Table 14.2 column 1, we indeed show that for families who are deemed to have shared physical custody arrangements, counting both parents' incomes is common practice. Nine of the thirteen countries adopt an income shares approach, with only three countries (Denmark, UK and Iceland) determining child maintenance liabilities based on the non-resident parent's income only. In the U.S. (Wisconsin) both parents' incomes are counted only in shared physical custody cases, not in sole physical custody cases where only the non-resident parent's income is assessed. At face value, when considering an incomes shares approach, it seems there is a greater recognition of gender equality in parental obligations post-separation when determining child maintenance amounts; at least that is when there is shared physical custody. However, we also need to consider whether the obligation to pay still exists. A system can theoretically use an incomes share assessment approachbut at the same time decide that there is no longer an obligation for either parent to pay child maintenance when it is deemed they have shared physical custody. Effectively, parents are considered to be taking equal responsibility, regardless of any disparities in their incomes. We explore that next in columns 2 and 3 of Table 14.2 where we show three possible outcomes: (a) that automatically no child maintenance is set because there is deemed to be shared physical custody (the child maintenance obligation is effectively annulled), (b) there is still an order made for child maintenance, but amounts may be adjusted/reduced, or (c) child maintenance is still required and no adjustments are made, meaning having shared physical custody makes no difference and parents pay the same amounts regardless.
Looking across columns 2 and 3 of Table 14.2, the results show that if there is equal time share only two countries (Denmark and the UK) deem there to be no obligation and therefore child maintenance is annulled. In France and Sweden it is annulled only if parents have equal incomes. Appositely, only in Estonia and Iceland do child maintenance guidelines not recognize the division of care as a factor that can modify child maintenance obligations: thus the child maintenance obligation is unchanged even in cases of shared physical custody. For the majority of countries however, an obligation to pay remains where there is equal time share and for most, the amounts are reduced/adjusted by varying degrees depending on the rules applied (Australia, NZ, Norway, Belgium, France, Spain (Catalonia), Sweden, Finland and U.S. (Wisconsin)).
In some of those countries however, there is a complex interrelationship between assessing care time and assessing the incomes of both parents in determining what the level of child maintenance should be. So it is not always the case that shared physical custody on its own reduces child maintenance, but rather an income discrepancy between parents may mean the richer parent still has to pay, despite having shared physical custody. We think that an income effect is operating in Australia, New Zealand, Norway, Belgium and U.S. (Wisconsin) (we discuss that further in the next section). In France and Sweden, we have recorded in column 3, Table 14.2 that child maintenance may be still be paid according to the rules, but it is unlikely this would happen in practice (even if parents have slightly different income levels) because very few of those who have shared physical custody received or paid child maintenance in France and Sweden, at least that was in 2004 and 2014 (Moreau et al. 2004;SCB 2014). Table 14.2 therefore only shows whether there is likely to be an effect as a result of shared physical custody, but not the actual child maintenance monetary outcomes that are produced. However, we can measure the strength of that effect when we examine the actual amounts of child maintenance calculated using the model families in the next section.

Levels of Child Maintenance
In this section we analyse the child maintenance schemes to show how much they would determine as being the formal child maintenance obligation in two different care-time scenarios. We calculate the amount the liable parent is obliged to pay per child in our fictitious model family (Fig. 14.1). We first show how much child maintenance would be set in each country for the base case which shows the first care time scenario; that is where children have two overnight stays every other weekend with one parent. In the next care time scenario, we analyse what happens when the children have shared physical custody arrangements (applying the 50/50 time threshold in each country). In all scenarios, we use the male and female median incomes for full-time earners and hold them constant. The amounts of child maintenance produced for the model family are reported in Fig. 14.1 and it assumes that it is always paid (of course in a real family this might not always be the case).
The first set of bars present the child maintenance amounts due in the first caretime scenario of two overnight stays per fortnight, and both parents have median incomes. In this situation, in all countries, the non-resident parent (the one who has the children to stay two nights per fortnight) is expected to pay child maintenance. The maintenance awards are clearly lowest in Sweden, followed by France and Belgium (less than 200 ppp$/month), while U.S. (Wisconsin), Estonia and Spain 4 (Catalonia), require the highest amounts (over 400ppp$/month).
The next set of bars show child maintenance liabilities when the model family moves from a situation of regular contact to one of shared physical custody (caretime scenario two: 50/50). Shared physical custody as tested in our model family has a greater impact on what the other parent is expected to pay. We cluster countries into three groups accordingly. In group 1, full reduction is taken of shared physical custody and a zero amount of child maintenance is set (full reduction). In group 2 a partial reduction is available and the maintenance amount is reduced to a greater or lesser extent across countries (partial reduction): in group 3, no account is taken of shared physical custody and therefore no reduction in child maintenance is made.
In the full reduction group 1, are four countries, Denmark, UK, France and Sweden and the child maintenance obligation would be set at zero. This reflects the assumption that if parents share care of their child equally (and for France and Sweden if the parents' incomes' are also roughly similar) then the cost of rearing the child must be met equally between them.
In the partial reduction group 2 are Australia, New Zealand, Norway, Spain (Catalonia) Belgium, Finland and the U.S. (Wisconsin). The policy and practice guidelines in these countries make more fine grained calculations of how much child maintenance should be paid in shared physical custody situations. When comparing the amounts calculated for our model family from having two overnight stays to having shared physical custody, then the level of reduction varies. Child maintenance amounts are reduced by at least a half if not more in Australia, New Zealand, Norway, Spain (Catalonia) and U.S. (Wisconsin), but reduced by less than a half in Belgium and Finland.
In the no reduction group 3, are Estonia and Iceland. In our model family, the other parent is still expected to pay the full amount of child maintenance even where there is shared physical custody and both parents work full-time. In those two countries the liable parent always pays the minimum payment, which seems to be a relatively high amount compared to the other countries, according to our calculations based on this model family using pppUS$.
Overall, again we see no obvious relationship between the type of child maintenance scheme and the three groups, other than to say that the only two counties that give a full reduction of maintenance irrespective of the parents' income are agency schemes Denmark and UK. That is because neither country uses an incomes shares approach to calculate obligations, also for the UK at least, the reason is to keep the administrative system simple.

Concluding Discussion
In this chapter, using new evidence from a comparative study we have filled the gaps in knowledge about how child maintenance schemes across different countries take account of shared physical custody arrangements. We have analysed data from 13 countries exploring this phenomena and have applied a model family approach that presents an idealised situation in which the parents in our model separated family are gender equal in terms of their work and care-time commitments.
We have found a very high degree of variation across countries and there is no obvious pattern in the approach adopted that relates to the type of child maintenance scheme in countrieswhether they are agency based, court based or a hybrid of the two. So neither the administrative rules nor judicial decision making in relation to different child maintenance schemes and their calculations show any clear consistency either within or across the scheme types. Still, it is somewhat surprising that we can find no relationship in our data as we have standardised our approach using model families. However, it is important to note that institutional and administrative arrangements do not fully explain the differences in child maintenance outcomes (Meyer and Skinner 2016). Even so, our findings in this regard could signify that internationally there is no communication about or consensus emerging on what the child maintenance obligations should be in the light of this phenomenon of more equal care arrangements being made between separated parents. Certainly, whilst the prevalence rates of joint physical custody may be growing (or are at least are believed to be growing) they are still not the common arrangement. That is according to our data provided by national informants' examination of the available administrative and survey evidence in their countries. An examination of the possible factors that might affect prevalence rates themselves (such as the availability of free childcare within countries) is beyond the scope of the research reported in this chapter.
What we have found however, is that most commonly, countries provide a partial reduction in child maintenance amounts in cases of shared physical custody compared to when the parents in the model family had the more typical arrangements whereby children spend two nights every other weekend living in one parent's household. Arguably, this more fine-grained approach could be considered better than the other two approaches (see below) as some account is taken of the gender pay gap, as this is what shows up in our model family where we use male and female median earnings. Of course this may not be an explicit policy intention underpinning child maintenance policies, but may simply reflect the application of operational procedures and judicial decision making based on judgments about each parents' capacity to pay. Even so, the outcome potentially creates a redistributive effect as the richer parent pays child maintenance to the poorer parent for the upkeep of the children.
Less commonly four countries provided a full reduction in child maintenance amounts when there was equal shared care. Thereby, assuming, that the situation between the parents in our model family was equal and therefore neither owed any child maintenance to the other. This approach has previously been criticized for two main reasons (see Melli and Brown 1994). First, it assumes that parents have similar incomes, which even in our idealised model family, is not the case given the gender pay gap in median earnings. In real life cases the picture is bound to be worse. Certainly, statistics show that mothers' total income decreases immediately after parental separation and very often is much lower than fathers' incomes. Indeed, fathers' incomes can even show a rise post separation (e.g. Andress et al. 2006;Mortelmans and Defever 2017). Second, this approach of providing a full reduction in child maintenance assumes that expenses are borne equally by both parents. However, not all costs relate to the time children spend living with parents as some of the childrearing costs may be paid disproportionally by one of the parents, irrespective of sharing care time. In reality, mothers often carry most of the responsibility for management of children's daily lives, including paying school-related expenses and health care costs (Cashmore et al. 2010). So whilst some country's child maintenance policies might be attempting to deliver equal treatment to both parents with similar time care and employment circumstances, the impact on outcomes might be anything but equal. Cook and Skinner (2018) point out that economically, for truly gender equal outcomes to be produced in separated families, equity based solutions might be needed that favour the more economically disadvantaged parent, which in societal terms are usually mothers. So in relation to our analysis, an equity solution would best fit with the fine grained partial reduction approach. However, regardless of which policy assumptions are in place, it is certainly a quicker and easier operational process to assume equality in family circumstances where there is shared physical custody, thereby avoiding calculating reductions in child maintenance amounts.
Finally, we found it was unusual to make no reductions in child maintenance amounts when there was shared physical custody (at least that is for our model family), It only happened in 2 of the 13 countries, Estonia and Iceland. Conceivably, the underlying operational assumptions in these countries could be based on a strong male breadwinner model in which the father is not excused from his economic obligation to pay full maintenance regardless of sharing care time.
Overall, it seems there is no standard practice in dealing with shared physical custody in child maintenance policies. The three different approaches that we found of making full reductions, partial reductions or no reductions in child maintenance amounts when there was equal care arrangements did not map easily onto the child maintenance typology, which highlighted different institutional settings. The latter is surprising, as it might have been expected that similar child maintenance schemes would treat shared physical custody in similar ways, or that one type of settingbe it court or agencywould show signs of being more responsive to perceived changes in social norms of shared physical custody as measured by the prevalence rates in countries. We found no evidence of institutional settings or prevalence rates having a key influence on child maintenance outcomes in our model family.
Many questions remain about recognising the sharing of care responsibilities between parents in separated families. There is patchy information on its prevalence and many interpretations of what it is and how to measure it across countries. This makes it not only very difficult to measure, but also difficult to consider a range of factors relating to other family policies (such as childcare provision) that might cast some light on the reasons for variations in rates across countries. More specifically, within child maintenance systems, it would be helpful to know more about the justifications underlying the different formulae used to measure shared care and the rationales of whether and how to make any adjustments/reductions in the amounts expected. Perhaps those institutions are the best places to investigate this phenomenon as they generally have to respond to separated parents' changing family practices and are therefore closest to understanding what is going on regarding shifting social norms around care arrangements.