About Nesta

Nesta is a research and innovation foundation. We apply our deep expertise in applied methods to design, test and scale solutions to some of the biggest challenges of our time, working across the innovation lifecycle.

What does the prior literature tell us about the potential impact of cash transfers for families with young children in the UK?

A large body of evidence suggests that growing up in poverty is harmful to child development and that increases in family income improve children's early school achievement.

In parallel, there is strong evidence that infancy and early childhood, especially the first 1,000 days of life, are key developmental periods, in which growth and development are most rapid and sensitive to change.

Bringing together these two bodies of research, Nesta believes that cash payments to families beginning early in life may be an important tool to improve developmental outcomes for children in disadvantaged families - helping us progress our 10-year mission to narrow the early attainment gap in the UK.

As an early step in testing the potential role of cash payments in narrowing the early education outcome gap in the UK, we set out to answer the question: what do we know from cash transfer trials about the impact of cash payments to families on early childhood development? To do this, we conducted a review of experimental evidence from high-income countries, focusing on studies that provide increases in income to parents of children who are under 5 years old.

We sought to identify one of two things, depending on the depth of the evidence base we found: either an estimate of the likely impact of cash payments in the UK, which would help us to inform further work, or an understanding of what evidence already exists and where the gaps are, which would be a starting point for further research, including a potential cash transfer trial.

Findings

  • There is limited experimental evidence of the impact of cash transfers on child development in high-income countries. Among eight trials reviewed, three were ongoing and four lacked published results. No experimental studies were found from the UK, and few from other high-income countries offer insights into the potential impact of substantial cash transfers to UK families.
  • Across studies with published findings, the developmental outcomes reported for children whose families received cash payments before age five are often marginally better than those in the control group. However, these differences do not reach statistical significance.
  • In addition, reporting across studies does not allow us to estimate the likely impact of cash payments in the UK on developmental or educational outcomes.
  • The three ongoing, US-based studies provide the best opportunity for estimates of the effect of cash payments on child development. However, findings may not directly translate to the UK, given differences in context.

Next steps

We at Nesta have already begun to build on the findings of this review by reviewing non-experimental, UK-based evidence for cash transfers to low-income families. We have summarised the results of this and discussed our current thinking in our report 'The impact of cash transfers: Investigating how money affects children’s development in the UK and other high-income countries'.

We will continue to pursue further work to better understand and design effective interventions to tackle the link between family income and child outcomes, including analysing available non-experimental data and tracking the identified highly relevant, ongoing studies.

As we delve deeper, we are keen to engage with different individuals or organisations also investigating and addressing the impact of income on families’ lives. We invite any researchers, policymakers, service providers, campaigning organisations, consumer and parent advocates, or others working in this space, both in the UK and globally to reach out to us at [email protected].

What does the prior literature tell us about the potential impact of cash transfers for families with young children in the UK?*

* The following text has been generated automatically from a PDF document. Please bear in mind that there may be some discrepancies between the original document and the automatically generated content. The original PDF is available to download and refer to.

What does the prior literature tell us about the potential impact of cash transfers for families with young children in the UK?

* The following text has been generated automatically from a PDF document. Please bear in mind that there may be some discrepancies between the original document and the automatically generated content. The original PDF is available to download and refer to.

--- ERROR CONVERTING CHUNK 1 ---

--- ERROR CONVERTING CHUNK 2 ---

--- ERROR CONVERTING CHUNK 3 ---

Domain Study ID Measure Child age at Assessment follow up period Result
Special education Child support demonstration evaluation10 Maternal report of child receiving special educational services. 6-11 years. During intervention. score difference was not significant. In 1998, 23.1% of children received special education services, and in 199 22.4%. Differences from the control group were small and non-significant.
School attendance Family rewards6 Both average attendance and % with attendance > 95% are reported. 8-11 years (2-7 years at time of randomisation). Six years post randomisation. Average attendance in intervention group was 85.8% compared to 85.1% in the control group, and no significant difference were seen in average or high attendance rates.
Domain Study ID Measure Child age at Assessment follow up period Result
Child support demonstration evaluation10 Maternal report that child missed 10+ days of school.1 6-12 years. During intervention. In 1998 8.4% of mothers reported their child missed 10 or more school days compared to 16% of control mothers and this difference was statistically significant (n=631, p=0.004). In 1999 there was a small and non-significant difference; 10% in experimental and 11.4% in control.

Table 4. Child health outcomes

Domain Study ID Measure Child age at assessment follow-up period Result
Health promotion activities Family Rewards11 % of children with a physical health examination within 12 months (parent report).2 < 8.5 years. 3.5 years post randomisation. Rate in intervention group 97.9%, in control 97.2%, not significant. This is a subgroup analysis (n=456).
% of children with a dental examination within 12 months (parent report).2 < 8.5 years. 3.5 years post randomisation. Rate in intervention group 59.2%, in control 57%, not significant. This is a subgroup analysis (n=456).
% of children received a screening (as recommended by a paediatrician (within 12 months) (parent report).2 < 8.5 years. 3.5 years post randomisation. Rate in intervention group 30.4%, in control 25.6%, not significant. This is a subgroup analysis (n=456).
Health Insurance Family Rewards5 % of families where all children covered by health insurance in the last month (parent report). Mixed age. 3.5 years post randomisation. Rate in the intervention group was 95.3, this was not significantly higher than the control group (Adjusted difference 1.4% CI -0.8-3.6%), although during the period of the intervention the rate was significantly higher (1.9% difference).
Child Support Demonstration Evaluation10 Maternal self report that their child1 was uninsured at some point in the last year. 0-5 years (subgroup). During intervention period. In 1998 14.2% of children were uninsured at some point, 1.5% lower than in the control group. In1999 this was 15.7% in the intervention group, 3.8% lower than in the control group. Neither difference were significant.
Sleep disturbances Baby's First Years12 PROMIS Sleep Disturbance-Short Form. Ages one, two and three. During intervention period. No significant differences observed.

Table 5. Child socio-emotional and behavioural outcomes

Domain Study ID Measure Child age at assessment follow-up period Result
Behavioural outcomes Child Development Accounts7 Social-emotional items from Ages & Stages11 Mean age 4.5 years. Treatment group had lower scores (indicating fewer behavioural problems) but not significant in the weighted analysis (after accounting for selection bias). However, this difference was significant in weighted analysis of planned subgroups of mothers living in rented accommodation, without a high school diploma, receiving welfare benefits and with a lower income.

Table 6. Parenting, family functioning and child maltreatment outcomes

Domain Study ID Measure Child age at assessment follow-up period Result
Child maltreatment Child support demonstration evaluation9 Record (in state child protection system) of child subject to a screened-in report for child maltreatment. Whole sample, so age not known, 60% of mothers had a child under three years. During intervention period. Children of mothers in the experimental group were less likely to have a child subject to a screened for maltreatment (odds ratio = 0.888, p<0.05).
Parenting Child Development Accounts8 Positive and punitive parenting practices assessed using a self-report tool. Four years. Some evidence of a modest effect of reduced punitive parenting practices, no effect on positive parenting.
Parenting quality Baby's First Years13 Parenting interactions with children: checklist of observations linked to outcomes.14 During intervention (approximately 12 months post baseline). No significant differences.
Economic pressure Child support demonstration evaluation10 Maternal self-report of reading to preschool child daily. 0-5 years. During intervention period. In 1998, 48.4% of mothers in the experimental group, 2.6% fewer than in the control. In 1999, 52.7% in the experimental group, 5.5% fewer than in the control group. Neither difference was significant.
Baby's First Years13 Food insecurity index – additive index of five items from the US Department of Agriculture's short-form measure of food insecurity.15 During intervention period (approximately 12 months post baseline). The impact of the treatment on food insecurity is positive, but not statistically significant after adjustment for multiple hypothesis testing (Westfall and Young adjustment), ITT estimate covariate adj. = 0.23, p (adjusted)= .15, p (unadjusted)=05, ES=.14.

Table 7. Parent outcomes

Domain Study ID Measure Follow up period Result
Maternal depression Child Development Accounts16 Maternal depression (Center for Epidemiologic Studies Depression Scale17). Four years. Mothers in the treatment group had significantly lower scores (1.83 vs 1.99, indicating less depression). The effect was small, but greater for lower income and lower education subgroups.
Baby's First Years13 Maternal depression (Personal Health Questionnaire Depression Scale PHQ-818). During intervention (approximately 12 months post baseline). No statistically significant differences (3.91 vs 3.71, ES=0.1).
Maternal anxiety Baby's First Years13 Maternal anxiety (Beck Anxiety Inventory). During intervention (approximately 12 months post baseline). Mothers in the high cash group had significantly higher levels of maternal anxiety (5.94) when compared with the low-cash group (4.58), p (adjusted)= 0.01, ES=.0.25 SD, N= 930.
Maternal stress/distress Family Rewards5 State of hope scale, average score. During intervention period (18 months post baselined). No significant differences (and very similar scores).
Family Rewards5 Kessler Psychological distress scales (19), both average score and % experience serious distress reported. During intervention period (18 months post baselined). No significant differences (and very similar scores) on both measures.
Baby's First Years13 Physiological stress (Ln Hair Cortisol) Perceived Stress Scale.20 During intervention (approximately 12 months post baseline). No statistically significant differences. (Ln Hair cortisol 1.89 vs 1.73 ES=0.02; PSS 11.39 vs 10.82, ES=0.1).
Substance use Baby's First Years4 Maternal alcohol use, cigarette use, and opioid use. Self-reported frequency scores. During intervention period (approximately 12 months post baseline). No statistically significant differences.
Spending estimated from maternal report household weekly purchases of alcoholic beverages and cigarettes. During intervention period (approximately 12 months post baseline). No statistically significant differences.

Table 8. Results of risk of bias assessments including rationale for identified sources of bias

Study Outcome RoB domains/sources of bias Assessed level of risk Rationale
Baby's First Years3 Maternal anxiety Measurement of outcomes. High Participants in the intervention group were most likely aware that they were receiving the intervention and a self-report measure of maternal anxiety was used. Early reporting of outcomes.
Family Rewards5 7 Selected reporting of results. Some concerns Slower-than-expected recruitment resulted in the programme beginning with half the sample. It is therefore unclear if the entirety of the intervention group received the intervention for the same duration of time.
English Language abilities at Year 6 Deviation from intended intervention. High Multiple domains of outcomes and time points are reported in the main report, and no primary outcomes or primary outcome points are pre-specified.
Child Support Evaluation9 Child welfare system involvement Selected reporting of results. Some concerns Researchers used mother's social security numbers (SSN) to assign participants to intervention or control groups. There are some concerns that, prior to 2011, SSNs were not assigned at random and that using them to assign to groups therefore introduces some bias (which may be observed in the difference in older maternal age in the intervention group).
Deviation from intended intervention. Some concerns Parents were aware they were receiving the intervention, and may have behaved differently as a result of knowing their child support was more likely to reach their child or be retained by the state. Bias may also introduced for other outcome measures that use self-report instead of routinely collected data.
Missing outcomes. Some concerns Researchers acknowledge that it is impossible to know whether some participants moved out of state and that any concerns recorded out of state would not be recorded. Thus, we cannot know how much the sample is affected by movements out of state, or by records of the outcome of interest not held within the state.
Selected reporting of results. Some concerns Some outcomes are reported for each year rather than across the two years of intervention.
Child Development Accounts7 Socioemotional development Selected reporting of results. Some concerns It is unclear as to whether the outcome at hand was part of a pre-specified analysis plan.

58 Victoria Embankment London EC4Y ODS +44 (0)20 7438 2500 [email protected] X @nesta_uk f nesta.uk www.nesta.org.uk ISBN: 978-1-916699-19-9

Nesta is a registered charity in England and Wales with company number 7706036 and charity number 1144091. Registered as a charity in Scotland number SCO42833. Registered office: 58 Victoria Embankment, London EC4Y ODS. CC BY NC SA



  1. Mothers were asked to answer survey questions about just one of their children. 

  2. These outcomes are available from health records during the intervention period, but as these were conditional activities we report here the post-intervention measure. 

  3. Baby's First Years 

  4. Baby's First Years 

  5. Family Rewards 

  6. Family rewards 

  7. Child Development Accounts 

  8. Child Development Accounts 

  9. Child support demonstration evaluation 

  10. Child support demonstration evaluation 

  11. Ages & Stages 

  12. Baby's First Years 

  13. Baby's First Years 

  14. outcomes. 

  15. food insecurity. 

  16. Child Development Accounts 

  17. Scale 

  18. PHQ-8 

  19. scales 

  20. Scale 

Authors

Patricia J Lucas

Patricia J Lucas

Patricia J Lucas

Principal Researcher, fairer start mission

Patricia joined the fairer start mission as a principal researcher, leading on experimentation and evidence.

View profile
Simran Motiani

Simran Motiani

Simran Motiani

Senior analyst, fairer start mission

Simran joins Nesta as a senior analyst for the fairer start mission.

View profile