USB v USA and another appeal [2020] SGCA 57

This case involves a pair of cross-appeals against the decision of the High Court judge (“the Judge”) in relation to the division of matrimonial assets in USA v USB [2019] SGHCF 5 (“GD”). This was a short marriage of five and a half years but it was preceded by a long period of cohabitation between the parties. This judgment is an opportunity to clarify the law relating to the identification of matrimonial assets and the application of the structured approach set out in ANJ v ANK [2015] 4 SLR 1043 (“ANJ v ANK”).

Background

The Husband and the Wife were married on 23 February 2011 and interim judgment of divorce was granted on 16 August 2016 (“the IJ date”). Prior to the marriage, they had cohabited for about 12 years. Although the Wife did not bear any children during the marriage, her two children from her earlier marriage lived with the couple from the time they began cohabiting, some 12 years before the date of the marriage.

The Husband is a lawyer and the Wife is a senior marketing director in a major real estate agency. There is no dispute that the Wife is the more financially astute spouse as she has made very successful real estate investment decisions. As at the IJ date, she owned 17 properties, some of which were held through companies in which she was the sole shareholder. The Husband, on the other hand, had not been doing well for some time prior to the IJ date. The assets within the pool of matrimonial assets were largely a result of the Wife’s successful financial planning.

The decision below

Before the Judge, it was common ground that eight of the 17 properties owned by the Wife must be classified as matrimonial assets. Of these eight, seven were purchased during the marriage and the eighth, though purchased before the parties married, was used as the matrimonial home. The parties, however, disagreed over whether the following assets should be included in the pool of matrimonial assets (“the Disputed Properties”):

(a) the Bedok North Property;

(b) Telok Blangah Property A;

(c) Telok Blangah Property B;

(d) the Compassvale Property;

(e) the Marina Boulevard Property;

(f) Robertson Quay Property A;

(g) Robertson Quay Property B;

(h) the Woodleigh Property; and

(i) the Leedon Property.

It was agreed that the Disputed Properties were purchased (but not fully paid for) before the marriage on 23 February 2011. The Judge decided that notwithstanding this, part of the value of each of the Disputed Properties should be included in the pool. This was on the basis that the “acquisition” of an asset under s 112(10) of the Women’s Charter (Cap 353, 2009 Rev Ed) (“the Charter”) refers not only to their “purchase”, but also includes the continuing process of payment through repayment of a mortgage (GD at [15]). We refer to the value so added to the pool in respect of such assets as their matrimonial pool value or “MP value” for short. As a result of a lack of evidence as to the precise sums paid during the marriage, the Judge applied different methods to determine the MP values of the Disputed Properties. We describe them below to the extent that they are relevant to the appeals before us.

For Telok Blangah Property A, Telok Blangah Property B and Robertson Quay Property A, to obtain the respective MP values, the Judge applied the formula , where amount paid towards the acquisition of each property during the marriage (ie, between the date of the marriage and the IJ Date); = total amount paid towards acquisition of each property as at the date of the ancillary matters hearing (“AM date”); and N = net value of the property (ie, market value less outstanding liabilities) as at the AM date (GD at [72]).

For the Leedon Property, the Judge observed that there was no information provided on the total outstanding liabilities on the property as at the date of the marriage (GD at [74]). However, there was evidence that the Wife had taken out a housing loan of $632,280 for the property around November 2011, ie, during the marriage. Given the lack of evidence showing how much the Wife had paid upfront for the Leedon Property, the Judge assumed that the remainder of the purchase price, apart from the loan of $632,280, had been paid before the marriage. This amounted to $31,843.61.

With respect to Robertson Quay Property B, the Judge found that it was not possible to determine the total amount applied during marriage by deducting the outstanding liabilities as of the AM Date from the original purchase price. This was because fresh liabilities had been incurred against the property after the IJ date. Hence, the Judge took the absolute amount which was paid towards the acquisition of the property during the marriage, namely, $53,439.16, the sum of the mortgage instalments, and added this as a “notional sum” into the pool (GD at [79]).

Parties agreed that the matrimonial home, the Sunrise Close Property, should be included in the pool (GD at [11] and [44]). As regards two loans taken out by the Wife in relation to the property, the Judge held that these were not fresh liabilities undertaken in anticipation of the divorce, but that they had been incurred several years before the end of the marriage (GD at [115] and [118]). Thus, the Judge refused to draw an adverse inference against the Wife (GD at [118]).

Having ascertained how much of the values of the Disputed Properties should be included in the matrimonial pool, the Judge valued the total pool at $9,626,759.63 (GD at [85]). Applying the framework set out in ANJ v ANK, the Judge found that the ratio was 5:95 and 25:75 (in favour of the Wife) for the direct and indirect contributions respectively between the parties. In determining the indirect contribution ratio, the Judge took into account the parties’ contributions during the 12 years of cohabitation prior to the marriage. In this regard, the Judge endorsed the view in JAF v JAE [2016] 3 SLR 717 (“JAF v JAE”) that pre-marital contributions may be a relevant factor in division (GD at [92]). Applying a 70% weightage to the direct contributions, the Judge found the overall ratio to be 11:89 in favour of the Wife.

In the proceedings below, both the Husband and the Wife urged the court to draw an adverse inference against the other spouse, but the Judge declined to draw one. Although he found that both parties had failed to provide full and frank disclosure, he also found that there was no substratum of evidence to establish a prima facie case against either party (GD at [116] and [117]). Furthermore, he had already included in the pool funds belonging to the Husband that had not been disclosed previously and the sale proceeds of the Wife’s properties.

Parties’ cases

In this appeal, the Husband challenges the Judge’s method of calculating the MP values of the Disputed Properties. In brief, the Husband’s position is that the Judge ought to have included the entire net value of each of the Disputed Properties, instead of merely including the sums that were applied to their acquisition during the marriage. He submits that the Wife treated all her properties as a single “investment portfolio”, borrowing against one property to finance another without any consideration of whether the properties were purchased before or after the marriage. Thus, the true value of the matrimonial properties would be reduced when they were used to refinance pre-marriage properties. Relatedly, this means that the values of the pre-marriage properties would correspondingly increase. As the Wife did not adduce evidence to explain how the loans against the matrimonial properties were used, an adverse inference should be drawn against her and the entire value of each of the Disputed Properties should be included.

The Husband also submits that it is incongruous to take into account the parties’ contributions during the period of cohabitation for the purpose of determining the ratio of division whilst disregarding these same contributions for the purpose of identifying the pool of matrimonial assets. Further, the Husband contends that the Judge used an incorrect method to calculate the values of Robertson Quay Property B and the Leedon Property.

The Husband further argues that the Judge erred in rating his indirect contributions at 25%. In this regard, the Judge failed to take into account, inter alia, the financial component of his direct contributions and his care of the two children during the cohabitation period. Furthermore, the Husband contends that the Judge was wrong to have assigned greater weight to the parties’ direct contributions. According to the Husband, the direct and indirect contributions should be given equal weight or the weightage should be, at most, 60:40 in favour of the parties’ direct contributions.

For her part, the Wife contends that the Judge erred by taking into account any portion of the values of the Disputed Properties. As the Disputed Properties were purchased before the marriage, they do not in any way constitute matrimonial assets. The Wife also disagrees with the ratio of division arrived at by the Judge. In particular, the Wife asserts that the Judge ought not to have taken the pre-marital contributions into account in determining the ratio of division. The correct indirect contribution ratio should thus be 15:85 in her favour, with direct contributions given a 90% weightage. As for financial contributions, the Wife submits that the Judge should have used the present valuation of the matrimonial assets in order to determine their ratio, as opposed to the parties’ historic contributions. This would produce a ratio of 2.61:97.39 in her favour.

Issues

The issues arising in these appeals relate to: (a) the identification of the assets in the matrimonial pool; (b) the appropriate method to determine the ratio of division; and (c) whether an adverse inference ought to be drawn against the parties.

The law on matrimonial assets

Identifying the pool of matrimonial assets

We first spell out the applicable principles governing the identification of matrimonial assets. Section 112(10) of the Charter provides:

Examining the above provision, it is clear that the intention of the legislature was to confine the court’s powers of division to assets relating to marriage. It is established law that in determining the legislative purpose of a provision, primacy must be accorded to both the text and statutory context of the legislation: Tan Cheng Bock v Attorney-General [2017] 2 SLR 850 at [43]. As a starting point, the title of s 112 of the Charter states that the provision deals with the “[p]ower of court to order division of matrimonial assets” [emphasis added], ie, assets relating to marriage. Further, s 112 falls under Part X of the Charter which, on the whole, pertains to matters relating to the dissolution of marriage. In the Charter, Parliament made the decision to confine the court’s power to divide assets belonging to divorcing parties to those they acquired during marriage. This was a matter of social policy, and one clearly within Parliament’s purview. The court cannot and should not wade into matters of social policy where the legislature has established a clear statutory framework providing for the division of assets relating to marriage. It is, thus, axiomatic that the court must disregard assets which were acquired during pre-marital cohabitation or during any non-marital relationship. The ownership of such assets would have to be determined in accordance with general property law principles. It should further be emphasised that in determining the length of a marriage, the court ought not to take into account the period during which parties were cohabiting. Marriage confers a legal status on the parties which carries with it specified rights and obligations. For non-Muslim couples who marry in Singapore, their marriage begins when they satisfy the various statutory prescriptions which the Charter sets out for the attainment of that status. The Charter governs all aspects of civil marriage and divorce in Singapore. Thus, under our law, it is inherently self-contradictory to treat parties as “married” when they were simply cohabiting.

Nonetheless, assets acquired before the marriage may still be subject to the court’s power of division if they are “transformed” into matrimonial assets. In this regard, at the end of a marriage, the assets that the parties own may be placed in up to four different asset categories. Section 112 of the Charter contemplates that assets in at least three categories may be subject to the court’s powers of division. The classes of assets that the parties may possess are:

(a) “Quintessential matrimonial assets” (to use a term first adopted by Justice Debbie Ong in TNC v TND [2016] 3 SLR 1172 at [40]): these are assets which either spouse derived from income earned during the marriage or to which either spouse or both spouses obtained legal title during the marriage by applying their own money, and the matrimonial home, whenever and however acquired. The entire value of these assets assessed as at the ancillary matters date (generally) will go into the pool.

(b) “Transformed matrimonial assets”: we use this term to denote assets which were acquired before the marriage by one spouse (or, more rarely, by both spouses), but which have been substantially improved during the marriage by the other spouse or by both spouses or which were ordinarily used or enjoyed by both parties or their children while residing together for purposes such as shelter, transport, household use, etc. Once transformed, the whole asset goes into the pool but if there is no transformation then, subject to (c) below, any asset acquired before the marriage even if acquired by both parties would be dealt with in accordance with general principles of property law.

(c) “Pre-marriage assets”: these are assets that either spouse acquired before the marriage and which the other spouse does not thereafter improve substantially or which are not used for family purposes. These stay out of the pool unless, as discussed below, they are partially paid for during the marriage by the owning spouse with income that would have been a quintessential matrimonial asset had it been saved up rather than expended on the pre-marriage asset. Then, the proportion of the value of the asset that was acquired during the marriage should go into the pool.

(d) “Gifts and inherited assets”: these assets whenever acquired by either spouse are not part of the pool unless transformed by substantial improvement or use as the matrimonial home. If transformed they should be treated in the same way as other transformed assets.

Even though assets in category (b) may have been acquired before the marriage, they may nonetheless be “transformed” into matrimonial assets if they meet certain statutory criteria. These statutory criteria create a nexus or link between the marriage and the assets in question such that they are transformed into matrimonial assets, despite the fact that they were acquired before the marriage. Category (b) assets may become transformed matrimonial assets because they have been substantially improved during the marriage, or through their use by the family as part of family life during the marriage.

We elaborate on the meaning of “substantial improvement” as applied to a category (b) asset. In our judgment, the reference to “substantial improvement” necessarily has an economic connotation. There are least two possible senses in which this phrase may be understood.

First, the improvement of such an asset must entail the investment of money or money’s worth for the improvement of the asset. The mere increase in the value of the asset does not mean that the asset has “improved”. In order for the asset to be transformed into a matrimonial asset, there must have been investment of some kind in the asset. The paradigm example would be renovation works performed on a residential or commercial property. These can easily be understood as increasing the sale value of such a property. However, even if the resale value does not increase because of market forces, a substantial renovation which makes a previously barely habitable home very much more comfortable or able to attract higher rental income could be considered a substantial improvement. Second, the improvement must arise from effort which can be understood as having economic value. For example, if the asset is a business belonging to one spouse, then development of the business by the other spouse or by both spouses during the marriage by sustained efforts could transform that asset into a matrimonial asset. In this regard, however, carrying out administrative or minor public relations activities or being a nominal director may not be sufficient. There should be an increase in turnover or in profitability or some other measurable improvement. It will always be a question of fact as to how the efforts of the non-owning spouse have contributed to an improvement in the asset. Ultimately, the court’s focus is on whether there has been some expenditure or application of effort towards the improvement of the asset (in an economic sense).

We draw attention to the requirement under s 112(10)(a)(ii) of the Charter that the improvement must come from the efforts of both spouses or the other spouse. If, for example, the husband acquired and fully paid for a house before marriage and the parties do not live in it and after marriage the husband renovates it, a plain reading of s 112(10)(a)(ii) of the Charter would exclude the house from the pool. In this regard, we consider that to the extent that the money spent on the renovation would itself have been a quintessential matrimonial asset, there would have to be some accounting of the same in the matrimonial pool but as this question does not arise in this case, we defer further exploration of the issue to an appropriate future time. Under s 106(5) of the Women’s Charter (Cap 353, 1985 Rev Ed), the predecessor provision to s 112, the requirement to expend effort was interpreted fairly liberally, such that a modest amount of effort was sufficient to satisfy the requirement of “joint efforts”. Notably, in Koh Kim Lan Angela v Choong Kian Haw and another appeal [1993] 3 SLR(R) 491, this court construed the wife’s attendance at her husband’s boutique business events as satisfying the requirement that she had jointly contributed towards the improvement of the value of the husband’s business. This generous interpretation of “improvement” may no longer hold true but that is a matter to be decided in an appropriate future case. It should also be appreciated that such efforts by a spouse can still be considered as part of his or her indirect financial contributions and be taken account of when it comes to the division exercise.

A category (b) asset can also be transformed if it is regularly used or enjoyed by the members of the family or for the benefit of the family. We are of the view that such use must be usual and relatively prolonged rather than casual. As an example a pre-owned saloon car which is regularly used to ferry family members around for activities like school and shopping and family outings would be transformed into a matrimonial asset whereas a pre-owned sports car which is generally driven only by the owning spouse with the children being taken out for a spin once in a blue moon would remain in category (c).

In the case of either substantial improvement or family usage, once the asset is transformed into a matrimonial asset, then the whole value of the asset will be included in the matrimonial pool for division, not just a portion thereof.

Next, we turn to the question of whether an increase in value of a pre-marriage asset (ie, an asset in category (c)) during marriage should fall outside the pool, given that such an increase does not constitute an “improvement” of the asset in the sense which we have described above. To be clear, in this section we are not referring to passive gains from market movements. Prof Leong Wai Kum suggests that a portion of the value of the matrimonial asset, which represents the value of the asset that has been “acquired” during the marriage, should be included in the pool. A number of High Court decisions, including the decision of the Judge, have followed this approach and treated an increase in value as “acquisition” of a portion of the asset in question. This requires us to consider how the term “acquired” should be interpreted.

In our judgment, the interpretation of the term should be approached sensibly in line with the overall legislative purpose evident from the provision as a whole. The starting point in the division exercise, as this court observed in BPC v BPB and another appeal [2019] 1 SLR 608 (“BPC v BPB”) at [50]–[51] (referring to Leong Wai Kum, Elements of Family Law in Singapore (LexisNexis, 3rd Ed, 2018) (“Family Law”) at para 17.065), is the identification of the material gains of the marital partnership:

An increase in value of an asset from payments made by the owning spouse during the marriage would represent the material gains of the marital partnership. It would be wrong to adopt a narrow or technical reading of “acquired”, such that “acquired” is interpreted to mean “purchased”. The court’s focus is on identifying the material gains of the marital partnership. It would not be right that an asset is excluded entirely from the pool simply because it was purchased a short time before marriage. The court should not allow the means by which an asset was acquired to detract from the fundamental purpose of the division exercise, which is to identify all the material gains of the marital partnership.

We give an example that should be familiar to all parties. It is extremely common that when a house or a similar asset of substantial value is purchased, a couple will take out a mortgage loan to pay for a substantial proportion of the purchase price. They will then pay off that mortgage loan during the marriage. When an asset has been purchased before marriage and the legal title of the asset passes before marriage, the court may nonetheless conclude that a portion of the value of the asset was “acquired” during the marriage. As the legal title had already passed before marriage, it is clear that what is “acquired” during the marriage cannot be the “legal title”. Instead, through the process of continuous repayment during the marriage of the loan secured by the mortgage on the property, what is “acquired” is the equivalent proportion of the equitable or beneficial interest of the mortgagee. In this paragraph we are referring to payments made by the spouse who purchased the property and holds the legal title. If substantial payments are made by the other spouse, then the entire asset would become a matrimonial asset as we discussed above.

On occasion, evidential difficulties may arise in proving the exact value of the portion of the asset that is acquired during the marriage. Where parties provide precise figures on the net values of an asset as at the date of the marriage and as at the date of the ancillary matters hearing, the court may easily ascertain the net increase in the value of the asset by calculating the difference. However, as was the case here, there may be a dearth of evidence on the net value of the asset as at the date of the marriage. This is often the case in respect of long marriages where an asset would have been purchased so many years earlier that the parties no longer have documents showing the value of the asset at the date of purchase or at the date of the marriage.

In our judgment, this evidentiary difficulty can be dealt with as a matter of the burden of proof. When a marriage is dissolved, in general all the parties’ assets will be treated as matrimonial assets unless a party is able to prove that any particular asset was either not acquired during the marriage or was acquired through gift or inheritance and is therefore not a matrimonial asset. The party who asserts that an asset is not a matrimonial asset or that only a part of its value should be included in the pool bears the burden of proving this on the balance of probabilities. This rule obviates many difficulties that may arise in the court’s fact-finding exercise and is consistent with the general approach to legal burdens in civil matters.

Conversely, we might add, where an asset is prima facie not a matrimonial asset, the burden would lie on the party asserting that it is a matrimonial asset to show how it was transformed. For example, in our recent decision in TQU v TQT [2020] SGCA 8, it was undisputed that a property at Pender Court was a gift from the husband’s father to the husband prior to the marriage (at [50]). The burden then fell on the wife to produce evidence that the property had been used as a matrimonial home and had therefore been transformed into a matrimonial asset, or that she had made substantial improvements to the property during the marriage (at [55]).

We observe, moreover, that such evidentiary issues more usually arise in the context of long marriages rather than in short ones, because in the latter the evidence is often readily available. Parties will not reliably document each spouse’s contribution or expenditure during a 20-year long marriage. Should there be real evidential difficulties in cases involving short marriages, this may suggest that parties were content to treat the asset as a matrimonial asset. This in turn would weigh in favour of any decision to include the full value of the asset in the pool.

Once the spouse has produced the necessary evidence, the question that arises is how the court should quantify the proportion of the asset that is to be included in the pool. One option is for the court to put into the pool only the amount spent after marriage, for example, the exact sum paid to reduce the mortgage loan. Another option is to apply a formula, similar to the approach applied by the Judge in relation to properties described above at [6], to determine the proportion of the current net value of the asset (which may be higher or lower than the amount spent) that should be credited to the pool. Generally, the latter approach may be preferred as it appears fairer and any capital gain would be reflected in the calculation, but we repeat our words of caution in UYQ v UYP [2020] 1 SLR 551 (“UYQ v UYP”) that parties should not take an overly mathematical approach. The particular approach adopted in each case will ultimately depend on the evidence and arguments put forward by the parties. The courts should adopt a common-sense approach to this calculation, and an appellate court will be slow to intervene with the judge’s exercise of discretion unless it is clearly wrong or inequitable.

A spouse who is able to show that the property was purchased before the marriage and can establish certain parameters, for example, that there was no real change in interest rates (in the mortgage) during the period in question, may be able to show that a proportion would have been paid prior to the marriage and might well persuade the court to apply a proportionate reduction of the value from the pool.

The approach to division

We recently affirmed the structured approach set out in ANJ v ANK in UYQ v UYP. Briefly stated, the structured approach prescribes the following steps: (a) first, ascribe a ratio that represents each party’s direct contributions relative to those of the other party, having regard to the amount of financial contribution each party has made towards the acquisition or improvement of the matrimonial assets; (b) second, ascribe a second ratio to represent each party’s indirect contribution to the well-being of the family relative to that of the other throughout the marriage; and (c) third, using each party’s respective direct and indirect percentage contributions, derive each party’s average percentage contribution to the family that would form the basis to divide the matrimonial assets (ANJ v ANK at [22]; see also BPC v BPB at [70]). In TNL v TNK and another appeal and another matter [2017] 1 SLR 609 (“TNL v TNK”), we held that the structured approach does not apply to long Single-Income marriages.

In our judgment, the structured approach should continue to apply to short marriages. In this regard, we reject the argument that the court should incline towards equality of division in short marriages. The court is inevitably constrained by the statutory language conferring its power to divide matrimonial assets under s 112 of the Charter. Section 112(2) states plainly that “[i]t shall be the duty of the court in deciding whether to exercise its powers under subsection (1) and, if so, in what manner, to have regard to all the circumstances of the case”. We repeat our observation in Lock Yeng Fun v Chua Hock Chye [2007] 3 SLR(R) 520 at [57]–[58]:

A key characteristic of the structured approach in ANJ v ANK is that it deals fairly with the two types of contributions – direct and indirect – whilst giving the court a discretion to set the weightage in terms of the relative importance of the direct and indirect contributions. This ability to vary the weightage between direct and indirect contributions is especially useful in a short marriage.

Further, unlike in a long marriage, parties in a short marriage would generally have in their possession evidence of their direct contributions to the marriage (see also the observation by this court in UYQ v UYP at [2] that it is an impossible exercise to take every detailed record of the marriage into account in the division exercise, especially in a long marriage). These direct contributions in a short marriage can be fairly taken into account by applying the structured approach in ANJ v ANK.

Generally, indirect contributions are less significant in short marriages, especially those without children. The court may take this into account by ascribing a higher weightage to direct contributions. In ATE v ATD and another appeal [2016] SGCA 2, this court assigned the parties’ direct contributions a higher weightage in a marriage lasting about five years. Andrew Phang JA held at [21]:

It should be reiterated that the court has a discretion to adjust the weightage of direct and indirect contributions in a short marriage – it is not compelled to arrive at an unequal weightage. There may well be good reasons for the court to retain equal weightage between direct and indirect contributions even in a short marriage, especially if there are children. Apart from short marriages, we add the important caveat that adjusting the weightage of the direct and indirect contributions should be done as an exception. This is because of the features of a long marriage as recognised by this court (ANJ v ANK at [27]):

Where the court exercises its discretion to adjust the weightage, apart from in a short marriage, we are of the view that cogent reasons should be provided to explain the departure from the norm of equal weightage.

In our judgment, the broad-brush approach should be applied with particular vigour in assessing the parties’ indirect contributions. This would serve the purpose of discouraging needless acrimony during the ancillary proceedings. Practically, this means that, in ascertaining the ratio of indirect contributions, the court should not focus unduly on the minutiae of family life. Instead, the court should direct its attention to broad factual indicators when determining the ratio of parties’ indirect contributions. These would include factors such as the length of the marriage, the number of children, and which party was the children’s primary caregiver.

The broad-brush approach towards determining the ratio of indirect contributions is also justifiable on the basis of another aspect of family proceedings, viz, the manner in which the veracity of evidence is tested in court. The evidence presented to the court in ancillary proceedings is usually in the form of affidavits, and the evidence is not tested by cross-examination. This point was made by the High Court in UDA v UDB and another [2018] 3 SLR 1433 at [38]–[39]:

In the absence of cross-examination, it is difficult to determine many of the matters evidencing a party’s indirect contributions to the marriage. For example, one parent may frequently pick his or her child up from school or guide the child in important but immeasurable ways. Such matters are neither documented nor subject to strict examination by the court, and rightly so. A marriage is not, and should not be, a cold and calculating commercial relationship, where parties keep a watchful eye on each other with a view to building up a case in court if they fall out (see also UYQ v UYP at [2]). It would be unwise to use the desire to determine veracity as a reason to liberally permit parties to cross-examine each other. Cross-examination tends to prolong and exacerbate the bitterness between the parties. It has long been recognised as part of the approach we take to family proceedings that, arising from the need to reduce rather than exacerbate tensions, courts will take a more flexible approach to evaluating the evidence. Often the evidence will not have been tested by cross-examination, and there will be much greater reliance on submissions and consideration of the surrounding facts and circumstances in order to arrive at findings and inferences that the court feels able to make. This does not change the burden of proof; rather it accommodates a less adversarial method of proof. One spouse’s failure to “prove”, in the strictest sense of the word, what he or she may have done during the marriage does not mean that a finding will be made against him or her on this point (see UYP v UYQ [2019] SGHCF 16 at [61] and UJF v UJG [2019] 3 SLR 178 (“UJF v UJG”) at [52]).

In light of the procedural constraints inherent in ancillary proceedings, each party’s duty of disclosure to the court takes on greater significance. The duty of full and frank disclosure underpins s 122 of the Charter (BG v BF [2007] 3 SLR(R) 233 at [52]). It is well established that the court is entitled to draw an adverse inference against a party who fails to comply with his or her duty of full and frank disclosure (BPC v BPB at [60], citing ANJ v ANK at [29] and NK v NL [2007] 3 SLR(R) 743 (“NK v NL”) at [57]). However, the mere absence of evidence is not sufficient to trigger an adverse inference; there must be a substratum of evidence that establishes a prima facie case against the party against whom the inference is to be drawn, and the party must have had some particular access to the information he or she is said to be hiding (UJF v UJG at [73]; BPC v BPB at [60]). Each party’s discovery obligations must be strictly observed. If they are unable to make the necessary disclosure, they must explain why; they cannot just ignore the obligation.

Taking cohabitation into account in the division exercise

We explained above that assets accumulated during cohabitation do not fall within the pool, save where they are transformed into matrimonial assets upon satisfying the statutory criteria under s 112 of the Charter. Relatedly, we stress that it would be wrong in principle for the court to take account of the parties’ indirect contributions during cohabitation when it is determining the extent of their contributions to the marriage.

High Court decisions, quite apart from the one currently on appeal, have adopted different approaches to this issue.

In UJF v UJG, the court declined to take “pre-marriage circumstances” into account in determining the extent of the parties’ contributions to the marriage. Aedit Abdullah J reasoned that there was nothing in the language of s 112 or the scheme of Part X of the Charter which justified taking into account such pre-marriage circumstances (at [54]). As the scope of s 112 only comprises factors relating to the period of marriage, the fact that the parties were in a relationship for many years before marriage was not a relevant factor to take into account in determining the ratio of division (at [54]).

Endorsing the contrary view, the court in JAF v JAE held that while the court did not have the power to divide pre-marital property, in determining the ratio of division it could take into account pre-marital contributions that enhanced the marriage (at [20]). The court took an opposing view from the court in UJF v UJG on the scope of s 112(2) of the Charter. The court observed that the ambit of s 112(2) is wide because it mandates the court to have regard to “all the circumstances of the case”. It was further reasoned that at least two of the listed factors under s 112(2) of the Charter allow the court to take into account factors which were unrestricted by time. There was therefore no general statutory restriction on taking into account matters which occurred before the marriage in determining the ratio of division under s 112 of the Charter.

Respectfully, we prefer the view espoused in UJF v UJG for the reasons contained in that decision. In our judgment, the phrase “all the circumstances of the case” must be read in light of the legislative purpose and overall statutory context of s 112 of the Charter. This directive under s 112(2) to take “all the circumstances of the case” into account does not give the court carte blanche to take account of matters that are unrelated to the parties’ marriage. A holistic analysis of the factors listed under s 112(2) shows that they are only concerned with circumstances relating to the marriage. Furthermore, as we pointed out above at [18], the statutory context of the provision makes it apparent that s 112 relates only to marriage. More fundamentally, taking into account indirect contributions made during cohabitation (and before marriage) would be inconsistent with our observation in NK v NL at [20] that “[t]he division of matrimonial assets under the [Charter] is founded on the prevailing ideology of marriage as an equal co-operative partnership of efforts” [emphasis added]. As a matter of principle, it would be wrong to count a party’s indirect contributions, albeit in arriving at the ratio of division, made during cohabitation when the power to divide is premised on the division of assets accumulated during marriage.

Our decision

Preliminarily, we reiterate the well-established principle that an appellate court will be slow to make minor adjustments to the orders made by the first instance court, and will seldom interfere in the orders made below unless it can be demonstrated that the court has committed an error of law or principle, or has failed to appreciate certain crucial facts: see TNL v TNK at [53] and BPC v BPC at [22].

Identifying and valuing the pool

As we stated above, the burden of proof is on the party claiming that an asset is not a matrimonial asset to convince the court.

We reject the Wife’s argument that the court should exclude the Disputed Properties entirely from the pool on the ground that they had been acquired prior to the marriage as legal title was acquired at the time. We have already explained the significance of payment of mortgage loans during the marriage. It is undisputed that the Wife continued to pay off the mortgage loans on the Disputed Properties during the marriage, and that is sufficient to bring the MP values of the Disputed Properties within the pool. The MP values of the Disputed Properties (ie, a percentage of their equitable or beneficial interests) were “acquired” during the marriage.

At the hearing, counsel for the Wife, Mr Koh Tien Hua (“Mr Koh”), sought to rely on s 51 of the Charter on the capacity of a married woman to hold property in her own name. That is entirely misguided. Section 51 was enacted to correct the historical common law doctrine of the “unity of legal personality”, under which a married woman lacked legal capacity and could not purchase property herself (see Family Law at para 3.020). That doctrine has long since been abolished in Singapore law. Marriage has no effect on a woman’s capacity to own property and married women hold property as feme sole. That provision was never intended to and does not affect the court’s power under s 112(10) to identify matrimonial assets for division upon divorce.

The duty of full and frank disclosure is particularly relevant in the context of ancillary proceedings. We do not think there is any reason to fault the Husband for failing to follow through on his summons for discovery. The duty of full and frank disclosure exists independently of applications for discovery and, especially in the context of matrimonial disputes, parties do not need an added incentive to apply for orders against one another. That having been said, the court will not draw an adverse inference against a party simply for non-disclosure of any asset. Parties must be reasonable in what they ask for. A party’s failure to comply with a summons for discovery is one factor that may weigh in favour of the court’s decision to draw an adverse inference against him or her.

In this case, initially the Wife was stubborn about providing disclosure. It is her own case that all the properties were used to fund one another. Upon the Husband’s request for discovery, the Wife provided the financing documents used for the properties purchased during the marriage and the matrimonial home. She refused to provide the documents for the other properties because she did not recognise them as matrimonial assets. It is, however, not for the Wife to say what assets do or do not belong to the pool and accordingly tailor the extent of disclosure. Ultimately, it is for the court, not the parties, to decide what belongs in the pool. Regardless of the parties’ subjective views on whether a particular property is a matrimonial asset, parties must assist the court to arrive at the correct decision by making full and frank disclosure. Otherwise, they bear the risk of an adverse inference being drawn against them. There was no good reason for the Wife not to have provided disclosure of the relevant documents for these assets. This was a short marriage so there was less time for memories to fade and for documents to go astray. Further, in all probability, she kept the necessary records, as evidenced by the fact that she subsequently produced the documents on the outstanding mortgage loan amounts for some of the pre-marriage properties.

Notwithstanding the Wife’s initial reticence, she eventually produced the relevant loan statements for the Disputed Properties. The Wife also produced a sheet tabulating her acquisition of the matrimonial properties. This table shows that, save for the Fraser Street property, she was still paying off the mortgages on those properties after the marriage ended. She had been making progress payments in relation to the Fraser Street property and had eventually forfeited the property to the developer; the amount refunded upon forfeiture was added to the pool by the Judge (GD at [50]).

The outstanding loan amounts on the properties are relevant because the crux of the Husband’s argument is that the Wife’s properties were managed as a single “investment portfolio”. The Husband submits that if the Wife took out a mortgage loan on a matrimonial property to finance one of the pre-marriage properties, that loan would logically have been taken out during the marriage, after the acquisition of the matrimonial property. That argument appears to us strange. It cannot be accepted for the following reasons.

First, it is far more likely in our view that at any point during the marriage the pre-marriage properties would carry lower loan burdens than the matrimonial properties did for the simple reason that the Wife would have had more time to pay off the indebtedness owing on the former. Second, even assuming that the Wife chose to re-mortgage a matrimonial property to pay off the mortgage loan of a pre-marriage property, there would be a corresponding increase in value of the pre-marriage property during the marriage, and this increase would already have been included in the Judge’s calculation of the value of the pre-marriage property. To accept the Husband’s submission would be to double-count that value. Similarly, to the extent that the Husband refers to rental income from the matrimonial properties being used to pay off the expenses incurred on the pre-marriage properties, that would already have been considered by the Judge as these payments would have decreased the debt burden on the pre-marriage properties during the marriage.

In his appellant’s case, the Husband was only able to highlight the loans the Wife took against the Sunrise Close property. The Sunrise Close property was purchased in 2003, prior to the marriage, but was included in the pool as it was the matrimonial home. The Husband claims that the Wife took out a loan on the property in 2008 and this would have been used to re-finance the other pre-marriage properties, and that the increase in value of the re-financed properties would not have been taken into account even though the new loan caused a decrease in the value of the Sunrise Close property. He uses this as an example to contend that the full values of the pre-marriage properties should be included in the pool. This argument is easily dismissed. The Sunrise Close property was a matrimonial asset only by virtue of its status as the matrimonial home. As a result, its full value was included in the pool notwithstanding the fact that it was acquired prior to the marriage. The Husband is correct that the value of the Sunrise Close property would have been reduced by the 2008 mortgage loan, and had the loan money then been used to pay down a mortgage on another property in 2008, the consequent increase in value of the other property would not have been included in the Judge’s calculations of contributions during the marriage. This would have been because all these adjustments would have taken place before the marriage and would not concern it. On the other hand, had any loan been taken against any property which had the status of being a matrimonial property because it was purchased during the marriage, such refinancing would also have occurred during the marriage; any reduction in value of that matrimonial property as a consequence would have been included in the Judge’s calculations, as would any corresponding increase in the value of the pre-marriage properties when the mortgage loans thereon were reduced or paid off. The 2008 loan against the Sunrise Close property and any consequence it had on the values of the other pre-marriage properties cannot be used to support the full values of all the Disputed Properties being included in the pool.

Second, it is apparent from the evidence that the 2008 “loan” valued at $545,000 was in fact an overdraft facility and not a term loan. That means the bank offered the option to the Wife to withdraw up to $545,000 on the security of the Sunrise Close property. There is no evidence, however, that she did in fact withdraw that sum and therefore no evidence of a decrease in the value of the Sunrise Close property in 2008.

To illustrate the point we made above, the Wife did in fact take out two loans against the Sunrise Close property during the marriage. These were taken in 2012 and in 2013. She explained that she refinanced the mortgage to enjoy lower interest rates and the loans “were utilised towards the purchase of other properties in [her] investment portfolio” and to pay for expenses including the mortgages and outgoings of various properties. If the loans were used to repay the mortgages on other properties, the repayments would have occurred in 2012 and 2013, ie, during the marriage, and that would have been included in the Judge’s calculations of the Wife’s contributions towards the pre-marriage properties during the marriage.

The only way that the value of the Sunrise Close property would have been diminished by the loans without a corresponding increase in the value of another property is if the Wife hid or dissipated the loan amount. That is not the Husband’s case. The 2012 and 2013 loans were taken before the breakdown of the marriage, and there is no suggestion that the Wife was siphoning off assets at that time.

The Husband relied on the Wife’s own statement that the properties were used to fund one another; he did not tender any independent evidence that the loans on the matrimonial properties were used to fund pre-marriage properties and, if so, to what extent. As we have explained, the management of the properties as a single “investment portfolio” would already have been included in the Judge’s assessment of the MP values of the Disputed Properties. The Wife’s statement on funding is not sufficient to establish a prima facie case against the Wife for the inclusion of the total values of the Disputed Properties in the pool. Contrary to the Husband’s assertion that the Wife failed to explain how the mortgage loans were used, she said, and it was sufficient for her to say, that she used them to pay off other mortgage loans and expenses. We do not think the Wife had to produce evidence of monthly repayments on the nine Disputed Properties, for example, or a trail of documents linking the mortgage loans to mortgage repayments. As we have already said, the parties must be reasonable in what they ask for. We highlight that the Husband does not suggest that she dissipated the loans. Even taking his case at its highest, the value of the loans would have already been included in the pool.

For the reasons stated above, we reject the Husband’s submission that the full values of the Disputed Properties should be included in the pool, and we do not disturb the Judge’s decision to include only a prorated portion of the values of the Disputed Properties.

Method of valuation

Admittedly, the Wife failed to disclose key pieces of information relating to several Disputed Properties. As a result, the Judge faced evidential difficulties in calculating the proportions of the following properties that were acquired during the marriage: the Leedon Property, the Marina Boulevard Property, the Woodleigh Property, Robertson Quay Property B and the Compassvale Property (GD at [74], [77], [78], [79] and [80]).

The consequences of the Wife’s non-disclosure were to a large extent rectified by the fact that the Judge painstakingly calculated the portion of each asset that belonged in the pool through the use of different valuation methods. We are in general agreement with the valuation methods that he adopted in light of the evidence that was before him, and we only make two observations in relation to Robertson Quay Property B and the Leedon Property.

First, the Wife incurred fresh liabilities against Robertson Quay Property B between the IJ date and AM date. To value other properties, including Robertson Quay Property A, the Judge used the formula (explained at [6] above). The Judge was of the view that he could not apply the formula to obtain the valuation of Robertson Quay Property B as he could not determine the correct net value as at the AM date due to the fresh liabilities incurred after the date of IJ. In the circumstances, the Judge took $53,439.16, which was the amount paid by the Wife in mortgage instalments throughout the course of the marriage, as the sum to be added into the pool.

We note that the Judge’s approach failed to take into account the capital appreciation of Robertson Quay Property B. We reiterate the point that there are various methods to place a value on the material gains of the marriage. Use of the formula is merely one of several different methods to calculate the value of the property acquired during the marriage. In our view, in general the Judge did not err by choosing one method over the other. However, his approach to Robertson Quay Property B was inconsistent with his approach to Robertson Quay Property A, which is located right next door. There is no reason why the net value of Robertson Quay Property B could not have been determined with reference to an earlier date, such as the IJ date. Given the original purchase price of $445,000 and the outstanding liability as at the IJ date of $200,812.70, it would appear that the Wife paid a total of $244,187.30 towards Robertson Quay Property B the prior to the IJ date. The amount paid during the marriage was $53,439.16. This was 21.9% of the total amount paid towards the property and therefore 21.9% of the value of the property as at the AM date, or $164,888.65, could have been added to the pool.

Second, the Leedon Property was purchased for $890,400. The option to purchase was dated 1 December 2010 and 20% of the purchase price was due within eight weeks, or by 26 January 2011. As the parties were married in February 2011, the Wife therefore paid at least $178,080 towards the property prior to the marriage. The next piece of available evidence shows that on 24 November 2011, after the marriage, the Wife took out a housing loan of $623,280 against the Leedon Property. The outstanding mortgage loan at the end of the marriage was $596,323.13. The Wife did not disclose how much she had paid towards the acquisition of the property between the date of the marriage in February 2011 and the date the housing loan was disbursed in 24 November 2011. The Judge assumed, in her favour, that the entire property had been paid for prior to the marriage. He included only 8.21% of the net value of the Leedon Property, or $31,843.61, in the matrimonial asset pool (GD at [74]).

The Wife bears the burden of proving what proportion of the property in question was not acquired during the marriage, which means the onus is on her to adduce evidence of the proportion paid before the marriage. This she has not done. In our view, the evidence suggests that a larger sum than the Judge found was paid towards the property during the marriage. The purchase price was $890,400 and 20% of the same amounting to $178,080 was paid in January 2011, leaving the outstanding balance of 80% or $712,000 to be paid. The housing loan taken out in November 2011 amounted to only $623,280 so the Wife must have paid another $80,040 between 26 January 2011 and 24 November 2011. Given that the parties were married on 23 February 2011, it is more likely that the sum of $80,040 was paid during the marriage. Adding the loan instalments paid during the marriage, the amount paid towards this property during the marriage would come to $106,996.87, or 32.6% of the sum paid towards this property overall. The amount that could have been added to the pool is 32.6% of the net value of the property of $387,863.69, or $126,443.56.

While the Judge went through the calculations to find the appropriate proportion of the value of each property to include in the pool, he did not have to do so. As we have said, the burden was on the Wife to provide the information pertaining to the Disputed Properties to prove the value that was acquired during the marriage. The court was justified in making a finding against the Wife as she failed to adduce evidence that was clearly in her hands and could have drawn an adverse inference against the Wife by, for example, increasing the Husband’s share of the matrimonial assets. That would have occurred after the Judge found the ratio of division, and we now turn to this exercise.

Ratio of division

The present case involves a short marriage and we had no difficulty in deciding to apply the structured approach. In determining the direct contribution ratio between the parties, the Judge adopted the broad-brush approach and concluded that the ratio was 95:5 in favour of the Wife (GD at [98]). Notwithstanding the sums that should be added to the matrimonial asset pool (see [71] and [73] above), we see no reason to disagree with the Judge’s conclusion and we affirm the application of the broad-brush approach. It was not in dispute that the Wife was solely responsible for the acquisition of the properties in her name and a ratio of 95:5 gives due regard to the parties’ respective direct contributions.

Turning to the indirect contribution ratio, as we have explained, the Judge should not have considered parties’ indirect contributions during the period of cohabitation. Still, we do not think the indirect contribution ratio should be adjusted for the simple reason that both parties’ indirect contributions to the marriage must be considered from the date of the marriage. That means that the Wife’s submission that the children had grown up by the time of the marriage, which we are in agreement with, applies with equal force to her contributions. Her efforts in raising them prior to 2011 must also be disregarded. By the time the parties married, the children were already 17 and 20 years old and would largely have been able to care for themselves.

In the circumstances and bearing in mind the broad-brush approach to ascertaining the indirect contribution ratio, there is no reason to disturb the ratio of 75:25 in favour of the Wife, which already fully recognises the Husband’s contributions to the family during the five years of the marriage. The Wife bore the burden of their financial support, including school fees and living expenses, and also appears to have cared for the children even though she was very busy.

As this was a short marriage, we also see no reason to disagree with the Judge’s decision to adjust the weightage between the direct and indirect contribution ratios, and we affirm his conclusion that the matrimonial assets should be divided in the ratio of 89:11 in favour of the Wife.

Conclusion

Given that we are not adjusting the ratio between the parties whether by reason of a different contribution ratio, a different weightage or an adverse inference, the issue is whether we should allow the Husband’s appeal on the two points on Robertson Quay Property B and the Leedon Property discussed above, which will lead to an increase in the value of the matrimonial asset pool of $206,049.44, or 2.14% from the original value of $9,626,759.63. The Husband’s 11% share of the assets, which was originally $1,058,943.56, will increase by a mere $22,665.44 or approximately 2%.

As we stated in TNL v TNK at [68], in the context of matrimonial disputes, appeals will not be sympathetically received where the result is a potential adjustment of the sums awarded below that works out to less than 10% thereof. We hardly think the sum of $22,665.44 justified an appeal before us. The approach in ANJ v ANK must be applied in a broad-brush fashion, and where the Judge below expended significant effort in particularising parties’ contributions and explaining his reasons for doing so, parties should not nit-pick at minor errors, especially where the errors are not ones of principle.

Given our observations on the appropriate valuations above, however, including the burden of proof and the importance of consistency, we allow the Husband’s appeal on these two points. The values of the Leedon Property and Robertson Quay Property B, of $31,843.61 and $53,439.16, shall be amended to the values of $126,443.56 and $164,888.65 respectively, and the total value of the matrimonial asset pool then becomes $9,832,809.07. The Husband’s 11% share is valued at $1,081,609. This means that the Wife will have to pay him $932,950.98 rather than the $910,285.44 ordered by the Judge.

Finally, we record our thanks to Professor Leong Wai Kum of the National University of Singapore’s Faculty of Law and Professor Chan Wing Cheong of Singapore Management University’s School of Law for their illuminating submissions as amici curiae.