A Difference February 2016 Dr Gordon Cleveland University of Toronto Scarborough Our ECE systems are marketbased and have parent fees Most early childhood education services are purchased by parents Services are supplied by private providers forprofit or notforprofit rather than pub ID: 563730
Download Presentation The PPT/PDF document "Affordability, Demand and Parental Emplo..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Affordability, Demand and Parental Employment: How Do Early Childhood Education Policies Make A Difference?
February 2016
Dr. Gordon Cleveland, University of Toronto ScarboroughSlide2
Our ECE systems are market-based and have parent fees
Most early childhood education services are purchased by parents. Services are supplied by private providers (for-profit or not-for-profit), rather than public.
The fees charged for early childhood influence parental decisions. Parents make decisions about employment, about type of care and particular provider.
Social objectives in ECE (e.g., employment, use, quality of education) are achieved through public policies that change provider decisions
and parental decisions.
Provider decisions are the
supply side
of the ECE market. Parental decisions are the
demand side
of the ECE market.Slide3
Therefore, ECE policies work through how they affect parental demand or provider supply
Demand will change when policies affect cost to parents, returns to working, or parental incomes.
Supply will change when policies affect costs of production, or fees that producers can charge.
Net effects of policy (and distributional effects) can be dramatically affected by changes in demand and supply
We have built a demand model for the City of Toronto to analyze the impact of different possible policy changesSlide4
Modelling Demand for Early Childhood EducationA team of us built a child care demand model for the City of Toronto last year
It models the child care decisions and the employment decisions of the main caregiving parent in families in TorontoSlide5
An Economic Model of Child Care Demand and Employment
T
heory: Model simplifies actual decision-making - Main caregiving parent chooses employment status and child care arrangements for all preschool children to maximize family well-being.
Parent can work full-time, work part-time, or be not currently employed.
Children can be in licensed child care, unregulated care by a non-relative and parent or relative care
(in Canada, most licensed child care is full-day and full-week) Slide6
1. Licensed child care/full time employment
2. Unregulated non-relative child care/full time employment
3. Parental or relative care/full time employment
4. Licensed child care/part time employment
5. Unregulated non-relative child care/part time employment
6. Parental or relative care/part time employment
7. Parental care/ main caregiving parent is not employed
Types of child care
Employment
StatusSlide7
Key factors affecting demand
Policy Levers - Affordability
Net price of child care
Main caregiving parent’s potential net income (incorporating effects of education, immigration status, cultural/ethnic background, single vs. two-parent)
Spouse’s net income or other net income
Background differences affecting decisions – may change over time
Age
of youngest child
Number of children 0-5 years
Main caregiving parent’s education
Single parent/two parent
Immigration status/timeCultural/ethnic background of main caregiving parentSlide8
An image of demand
Imagine there are 2 families. One has a single mother, born in Canada, of
British
ancestry,
with one child 3 years of age, and the mother has a university degree and a high potential wage. Based on these characteristics, she will have a high probability of being employed and using licensed child care.
The other family is two-parent, with the mother born in India and has recently immigrated to Canada. There are two children – 1 and 5 years of age. The mother has a high school education and a relatively low potential wage. Her husband makes a reasonably good income working as an engineer. Based on these characteristics, she will have a high probability of not being employed, but if she is employed, she will probably have a part-time job and provide most of the child care by herself, with help from relatives or her husband (by rearranging their shifts to be available for child care).
When Early Childhood Education policy changes, it usually affects either parent fees for child care or the net income of the main caregiving parent or other income in the family. This changes the probability that each family will use licensed child care, but
…
IT AFFECTS EVERY FAMILY DIFFERENTLY, BECAUSE EVERY FAMILY STARTS IN A DIFFERENT PLACESlide9
What’s the use of a demand model?
To forecast future demand for licensed services and future employment patterns
, even if policy does not change (driven by changes in total population, family incomes, immigration patterns, education levels, single parent vs. two parent families, cultural and ethnic background of families, numbers of multiple-child families etc.);
To predict the effect of ECE policies
that change any of these: parent fees for child care, net wages of main caregiving parents, other family income, taxes and benefits related to children (and convenience or quality of services);
And, to explicitly model the
interaction among different policies
To predict the “distributional” effects of policy
– who will benefit, by how much, who will be first in line to access scarce services;
To know what alternatives families are using
, where they go when licensed services become less accessible or affordable, and from which child care arrangements they switched when licensed services become more accessible and affordable;
To measure the “affordability” of licensed services and assess the changes in affordability when policies change.Slide10
1. To forecast future demand for licensed services and future employment
patterns
Current Supply and Estimated Demand for Licensed Child Care Spaces
2015
Current Supply
Demand
Difference
Licensed spaces (0-5)
47,136
51,205
4,069
(Percent of population)*(28)
(31)
(3)
Infant
3,311
3,710
399
Toddler
9,087
10,510
1,423
Preschool
22,769
23,415
646
Kindergarten
11,969
13,570
1,601
Licensed spaces (6-9)
18,728
21,975
3,247
(Percent of population)*
(17)
(20)
(3)
Total licensed spaces
65,864
73,180
7,316
(Percent of population)*
(24)
(27)
(3)Slide11
2. To predict the effect of ECE policiesSIMULATIONS:#1
- Funding is available to provide child care subsidies to all eligible families
#2 - No family pays more than 10% of net family income for licensed child care
#3 – The price of licensed child care is capped at $20 per daySlide12
Effects of Three Policy Simulations
Current Use
Simulation 1: Subsidies for all eligible families
Simulation 2:
Cap of 10% of family income
Simulation 3:
Cap of $20/day per child
Children 0-5 years
47,136
76,13579,77586,625
Additional demand
--
+ 28,999
+ 32,639
+ 39,489
(Percent of population)
28%
45%
(+17pp)
48%
(+ 20pp)
56%
(+ 24pp)
Numbers of main caregiving parents
Employment
87,270
93,915
94,180
96,115
FT employment
60,705
73,600
75,790
79,260
PT employment
26,565
20,315
18,390
16,885Slide13
Interactions among policies
All income taxes, employment insurance and compulsory pension contributions are calculated in model for each family.
All child-related benefits and child-tax-benefits are calculated for each family.
Child Care Expense Deduction is much less regressive than expected due to interactions.
Calculating distributional effects requires including taxes, benefits and child care policiesSlide14
3. To predict the distributional effects of policy
Projected Use of
Licensed Child Care by Household Income
Expected Household Annual Income (Before Tax)
Probability of using licensed child care
Base Case
Simulation 1:
Targeted subsidies
for all eligible families
Simulation 2: Cap of
10%
of
family income
Simulation
3: Operating Grant to reduce fees to
$20/day per child
Less than $50,000
38%
62%
56%
54%
$50,000 - $99,999
29%
41%
45%
50%
$100,000 or more
33%
34%
44%
59%Slide15
Distribution by Parent’s Education
Education Level of Main Caregiving
Parent
Probability of using licensed child care
Base Case
Simulation 1:
Targeted subsidies
for all eligible families
Simulation 2: Cap of
10%
of
family income
Simulation
3: Operating Grant to reduce fees to
$20/day per child
High School
28%
45%
43%
44%
College
35%
48%
51%
57%
University
36%
45%
51%
61%
33%
47%
49%
54%Slide16
Average
Probability of Using Licensed Child Care
in City of Toronto 2015
Youngest child age 1
16%
Youngest child age 3
66%
Youngest child age 5
39%
Recent immigrant – less than 5 years
27%
Canadian-born
37%
One child 0-5
37%
Two children 0-5
24%
High school education or less
28%
University degree
36%
Sole parent families
53%
Two parent families
29%
Average of all families
33%
Family characteristics matterSlide17
4. To know what child care arrangements families are using when they don’t use licensed services
The diagram on the next page shows the use of four main child care/employment arrangements that families use
In particular, the diagram shows how the use of each one is affected by a change in the net parent fee for the use of licensed careSlide18Slide19Slide20
5. To measure the affordability of licensed services
Two measures of affordability
for each family in the model
.
Family Income Affordability Measure (FIAM) – Net cost of child care to parents ➗ net family income
Caregiving Parent Affordability Measure (CPAM) – Net cost of child care to parents ➗ increase in net family income when caregiving parent is employed rather than not employed
CPAM measures the size of the employment “barrier” due to child care costs.
Note that caregiving parent’s income is the income that could be earned if she were employed full-time. “Projected” income.Slide21Slide22Slide23Slide24
Average value of Family Income Affordability Measure for this group (%)
Average value of Caregiving Parent Affordability Measure for this group (%)
Youngest child age 1
33.0
87.7
Youngest child age 5
10.0
22.5
Recent immigrant – less than 5 years
32.0
82.0
Canadian-born
20.5
52.2
One child 0-5
21.1
51.5
Two children 0-5
32.5
93.0
High school education or less
30.4
75.1
University degree
19.1
49.1
Sole parent families
26.7
42.9
Two parent families
24.1
68.0
Average of all families24.663.4AVERAGE VALUES OF AFFORDABILITY FOR TORONTO FAMILIES, BREAKDOWN BY SUB-GROUPSSlide25
Effects on Affordability: Three Alternative Policy Simulations
Degree of Affordability
Actual
%
Simulation 1: Subsidies for all eligible families %
Simulation 2: Cap of 10% of family income %
Simulation 3: Cap of $20/day per child
%
Affordable (<10% of net family income)
25
60.5
10076Unaffordable (10%-20% of net family income)
25
27
0
20
Completely Unaffordable (>20% of net family income)
50
12.5
0
4Slide26
Key Price Elasticities
Description
Size of elasticity
Details
Elasticity of Licensed Child Care Demand to Price of Licensed Care
- 1.14
Across all observations
Elasticity of Full-time Employment and Licensed Child Care to Price of Licensed Care
- 0.32
Measured at means of explanatory variables
Elasticity of Employment and Licensed Child Care to Price of Licensed Care
- 0.14Measured at means of explanatory variables
Elasticity of Parent and Relative Care to Price of Licensed Care
0.35
Across all observationsSlide27Slide28
Conclusions/Summary
Having statistical models to estimate demand, employment and affordability improves our ability to think about policy impacts
This type of model is excellent as the starting point for measuring the costs and benefits of possible policy changes
The demand model helps us understand the substitutions families make when affordability changes
Affordability of child care is a major barrier to the use of licensed child care and to full-time employment