Economic Insights
Financial Expectations and Household Debt

This Economic Insights article quantifies the degree to which families who expect their financial situation to get better in the next two years have, all else equal, more debt than comparable families. The data are drawn from the Survey of Financial Security for 1999, 2005 and 2016. The term “family” is used to refer to family units and includes economic families and unattached individuals. Debt and income estimates are shown in 2016 dollars. The study shows that even after a large set of socioeconomic characteristics is controlled for, families who expect their financial situation to improve in the near future have significantly more debt and generally higher debt-to-income ratios than other families.

Introduction

Economists usually agree that workers, families and firms incorporate expectations into their decision-making process.^Note Workers who expect their employer to implement mass layoffs in the near future may, as a precautionary measure, start searching for a new job, leave the firm or move into self-employment. Consumers who anticipate increases in durable goods prices may purchase products sooner. Firms that expect increasing demand for their products may choose to invest in machinery and equipment to increase their production capacity. Likewise, families who expect their financial situation to improve in the near future (e.g., because the major income earner anticipates a promotion) may increase their consumption expenditures, therefore carrying greater amounts of non-mortgage debt than they would have otherwise, or may choose to move to a larger dwelling and carry a higher mortgage.

It is generally difficult to assess how much expectations matter empirically since statistical agencies do not often collect this information. Fortunately, there are exceptions. In 1999, 2005 and 2016, Statistics Canada’s Survey of Financial Security (SFS)—a survey of families’ assets and debts—asked Canadian families the following question (Survey of Financial Security n.d., Question K17):

“In the next two years do you think your (family’s) financial situation will get better, worse, or stay the same?”

This question allows researchers, inter alia, to analyze whether families who expect their financial situation to improve carry significantly more debt than otherwise comparable families. These analyses may shed light on factors that underlie the evolution of household indebtedness. This issue is particularly interesting to policy makers and Canadians given the secular increase in various measures of household indebtedness, such as the debt-to-disposable income ratio of Canadian households (Chart 1).

Data table for Chart 1



Data table for chart 1
Table summary
This table displays the results of Data table for chart 1. The information is grouped by Quarter and year (appearing as row headers), Ratio, calculated using percent units of measure (appearing as column headers).

Quarter and year	Ratio
	percent
1990
Q1	85.42
Q2	87.47
Q3	87.41
Q4	87.25
1991
Q1	86.57
Q2	87.61
Q3	88.26
Q4	88.01
1992
Q1	88.19
Q2	88.81
Q3	88.78
Q4	90.08
1993
Q1	89.03
Q2	89.76
Q3	90.58
Q4	90.80
1994
Q1	90.89
Q2	92.38
Q3	93.26
Q4	94.08
1995
Q1	93.47
Q2	93.76
Q3	94.93
Q4	95.06
1996
Q1	95.35
Q2	96.70
Q3	97.31
Q4	97.90
1997
Q1	98.70
Q2	99.32
Q3	101.20
Q4	101.67
1998
Q1	102.87
Q2	103.20
Q3	104.76
Q4	104.55
1999
Q1	104.97
Q2	105.65
Q3	106.25
Q4	106.44
2000
Q1	106.52
Q2	106.88
Q3	107.22
Q4	106.54
2001
Q1	105.30
Q2	106.35
Q3	107.47
Q4	108.38
2002
Q1	109.12
Q2	110.44
Q3	111.81
Q4	112.60
2003
Q1	112.80
Q2	113.71
Q3	115.94
Q4	117.71
2004
Q1	118.20
Q2	120.84
Q3	122.36
Q4	123.65
2005
Q1	125.25
Q2	127.74
Q3	129.99
Q4	131.53
2006
Q1	131.36
Q2	133.23
Q3	135.21
Q4	136.45
2007
Q1	137.56
Q2	140.79
Q3	142.98
Q4	144.49
2008
Q1	146.31
Q2	148.65
Q3	149.47
Q4	149.60
2009
Q1	149.78
Q2	153.03
Q3	155.76
Q4	157.91
2010
Q1	156.59
Q2	158.58
Q3	159.51
Q4	159.91
2011
Q1	158.80
Q2	160.07
Q3	160.64
Q4	161.15
2012
Q1	160.15
Q2	161.43
Q3	161.95
Q4	161.61
2013
Q1	161.16
Q2	161.89
Q3	162.47
Q4	162.20
2014
Q1	161.80
Q2	162.78
Q3	164.05
Q4	163.98
2015
Q1	163.05
Q2	164.44
Q3	165.56
Q4	166.29
2016
Q1	168.09
Q2	171.68
Q3	173.75
Q4	174.90
2017
Q1	174.26
Q2	175.75
Q3	176.02
Q4	175.50
2018
Q1	173.94
Q2	174.86
Q3	175.82
Q4	176.28
Source: Statistics Canada, Table 38-10-0235-01 Financial indicators of households and non profit-institutions serving households, national balance sheet accounts.

Using the SFS for 1999, 2005 and 2016, this article quantifies the degree to which families who expect their financial situation to improve in the next two years have, all else equal, more debt and higher debt-to-income ratios than observationally equivalent families.^Note

Expectations by age

Since individual earnings tend to increase faster at the beginning of a career than during subsequent stages, one would expect young families to anticipate stronger growth in employment income and thus to be more optimistic about their financial situation in the near future than older families. Chart 2 supports this view. Regardless of whether families carry debts or non-mortgage debts, the percentage of families with positive expectations for their financial situation in the near future declines monotonically with the age of the major income earner. Specifically, families where the major income earner is younger than age 35 are at least 3.5 times more likely than families where the major income earner is aged 65 or older to expect their financial situation to get better in the next two years.

Chart 2 also shows that, within a given age group, families with some debt report slightly more optimistic expectations than other families. One interpretation is that families carrying relatively large amounts of debt might be more likely than others to pay off some of this debt quickly and would, therefore, be more likely to expect their financial situation to improve based on the resulting balance sheet realignment. It is not possible to test this hypothesis with SFS data. Nevertheless, the fact that younger families report more optimistic expectations than older families, regardless of their debt status, is consistent with the notion that their expectations for their financial situation contain information about expected growth in employment income.^Note

Data table for Chart 2



Data table for chart 2
Table summary
This table displays the results of Data table for chart 2. The information is grouped by Age group of major income earner (appearing as row headers), Without debts, With debts, Without non-mortgage debts and With non-mortgage debts, calculated using percent units of measure (appearing as column headers).

Age group of major income earner	Without debts	With debts	Without non-mortgage debts	With non-mortgage debts
	percent
Younger than 35	62.2	68.9	60.3	70.1
35 to 44	49.9	57.1	49.0	58.6
45 to 54	39.6	48.5	41.7	48.8
55 to 64	20.8	38.0	23.2	38.6
65 and older	8.5	19.8	9.0	20.1
Source: Statistics Canada, Survey of Financial Security, 1999, 2005 and 2016.

Expectations and non-mortgage debt

Table 1 assesses whether families who have positive expectations carry greater amounts of non-mortgage debt than other comparable families. Data from 1999, 2005 and 2016 are pooled to perform multivariate analyses. These analyses indicate that even after controlling for a large set of covariates,^Note homeowner families with positive expectations held roughly $6,800 (2016 dollars) more in non-mortgage debt than other homeowner families during the period from 1999 to 2016. Since the average non-mortgage debt among homeowners (including families with no non-mortgage debt) amounted to about $22,300 during this period, the difference is empirically significant: it represents 30% of average non-mortgage debt for this group.

Among homeowners, positive expectations for their financial situation in the near future are associated with significantly greater amounts of non-mortgage debt, even when the data are disaggregated by the age, education level or immigration status of the family’s major income earner. The same is true for renters, although the differences for this group are often estimated less precisely.

In sum, Table 1 indicates that families who have positive expectations about their financial situation in the near future generally have more non-mortgage debt in a given year than other comparable families.



Expectations and the debt-to-income ratio

Families with positive expectations about their financial situation may not only incur greater expenditures and non-mortgage debt than other families, but they may also choose to buy bigger homes and carry larger mortgages. This could happen if, for example, the family expected its employment income to permanently increase over the next few years.

Multivariate analyses support this hypothesis. All else being equal, homeowner families with positive expectations for their financial situation held about $27,900 more in mortgage debt than other families during the period from 1999 to 2016.^Note Since the average mortgage debt of homeowners (including those with no mortgage debt) amounted to about $101,000 during this period, the difference represents 28% of the average mortgage debt for homeowner families.

Therefore, homeowner families who expect their financial situation to improve in the near future display higher amounts of non-mortgage debt and mortgage debt than other families. This suggests that homeowner families with positive expectations will have greater debt-to-income ratios than other families.

Table 2 examines this issue for families whose after-tax income equals $10,000 or more (2016 dollars) and who are in the bottom 95% of the debt-to-income ratio distribution for the years 1999, 2005 and 2016, collectively.^Note For this sample, the debt-to-income ratios of homeowner families with positive expectations were, on average, 32 percentage points higher during the period from 1999 to 2016 than the debt-to-income ratios of other comparable families.^Note This difference is substantial since the average debt-to-income ratio observed for the sample of homeowners used in Table 2 amounted to 117% during the 1999-to-2016 period.

Regardless of the age group, education level or immigration status of the major income earner, homeowner families who expect improvements in their financial situation have debt-to-income ratios exceeding those of other homeowner families by at least 27 percentage points.

Among renter families, those who have positive expectations also have, in the aggregate, higher debt-to-income ratios than renter families with other expectations. However, the differences are less pronounced, at least in absolute terms. Differences are precisely estimated for renter families whose major income earner is Canadian-born, 35 or older, or has no university degree.

Overall, Tables 1 and 2 provide evidence that, in a given year, families’ expectations about their financial situation in the near future are strongly correlated with their levels of indebtedness.

This does not imply, however, that changes over time in families’ expectations can account for the increase in household indebtedness observed from 1999 to 2016. Since the percentage of families with positive expectations changed little during this period—from 46% in 1999 to 44% in 2016—changes in expectations cannot account for the observed increase in household indebtedness. Therefore, this suggests that other unobserved factors—e.g., a preference for bigger houses, a desire to maintain status with rich households in a context of rising income inequality (Bertrand and Morse 2016) or a falling aversion to high levels of indebtedness—may underlie the growth in household indebtedness since the late 1990s.



Conclusion

This article shows that families’ expectations about their financial situation are, in a given year, strongly correlated with their amounts of debt and debt-to-income ratios. All else equal, families who expect their financial situation to improve in the near future have significantly higher levels of debt and debt-to-income ratios than comparable families.

One important issue is the degree to which these correlations represent a causal effect of expectations regarding income growth on household indebtedness or some unobserved trait (such as optimism) that leads families to simultaneously expect improvements in their financial situation and carry large amounts of debt. Without panel data on wealth and expectations, it is impossible to disentangle these two mechanisms. Another possibility is that the positive correlation between expectations and indebtedness arises partly because families who carry relatively large amounts of debt might be more likely than others to pay off some of these debts quickly, and therefore expect their financial situation to improve. A third possibility is that past changes in a family’s financial situation might drive both its expectations and its current debt levels. Regardless of the underlying mechanisms generating this correlation, the results shown in this article highlight the fact that families’ expectations for their financial situation are strong predictors of household indebtedness in a given year. This finding is of interest to statistical agencies and researchers. It suggests that the collection of data on families’ expectations enriches analyses of household indebtedness and that such analyses could routinely include expectations as a control variable.

References

Bertrand, M., and A. Morse. 2016. “Trickle-down consumption.” Review of Economics and Statistics 98 (5): 863–879.

Lucas, R.E. 1976. Econometric policy evaluation: A critique. Carnegie-Rochester Conference Series on Public Policy, Vol. 1. Amsterdam: North Holland.

Statistics Canada, n.d. Survey of Financial Security – 2005 (SFS). Last updated November 22, 2012. Available at: http://www23.statcan.gc.ca/imdb/p3Instr.pl?Function=getInstrumentList&Item_Id=32170&UL=IV& (accessed February 6, 2019).