Assessing the Silver Support Scheme and policy implications
Agarwal, Qian, Ruan and Yeung (2022) investigate how recipients respond to the additional income from the scheme, and discuss the policy implications of their findings.
21 December 2022
For most of human history, the average lifespan remained well below 50 years. Today, however, the problem of widespread premature death is rendered moot as many countries worldwide grapple with ageing populations. Amid improving healthcare and declining fertility rates, the global population aged 60 years or over numbered 962 million in 2017, and is expected to double by 2050 according estimates by the United Nations. The sheer size of this demographic segment – compounded by the fact that many do not have sufficient retirement savings – means that governments have act for the elderly.
Deep diving into Silver Support Scheme
Announced by Singapore’s Prime Minister Lee Hsien Loong in 2014 and rolled out in July 2016, the Silver Support Scheme is a means-tested subsidy programme designed to help the bottom 20% of elderly individuals by distributing a quarterly cash subsidy to eligible Singaporeans aged 65 and above. The original edibility criteria included: (i) recipient’s pension savings (i.e. CPF) did not exceed S$70,000 by age 55; (ii) the monthly income per person in the household did not exceed S$1,100; and (iii) residential status which, among other things, required the recipient to live in public housing not larger than a 5-room flat.1
In their study “Supporting seniors: How low-income elderly individuals respond to a retirement support program”, Agarwal, Qian, Ruan and Yeung (2022) investigate how subsidy recipients use the additional income, and discuss the policy implications of their findings.
Empirical design
Agarwal et al.’s research compares the consumption behaviours of subsidy recipients before and after programme inception. By doing so, the study measures the recipients’ marginal propensity to consume (MPC), which is interpreted as the amount spent on consumption of goods and services per dollar of subsidy received.
Customer data from DBS Bank, Singapore’s largest bank, were used for sampling. The quarterly Silver Support subsidies are recorded with a unique transaction code which allowed the researchers to accurately identify subsidy recipients among elderly individuals. After applying subsidy eligibility criteria on transactional data ranging over a 36-month period from January 2016 to December 2018, the final sample comprised 1,340 individuals.
Findings
Regression analysis applied on the sample of 1,340 individuals showed that, on average, Silver Support recipients increased their total spending by 0.73 dollars per dollar of subsidy received, with 80% of the overall MPC attributable to cash spending (0.58 dollars per dollar of subsidy received). By analysing the locations of ATMs from which recipients withdrew cash, the researchers found that recipients expanded their geographic footprint and increase their dining-out spending, as proxied for by ATM withdrawals near food courts.
Though the recurring Silver Support subsidies represented a permanent increase in income for recipients, most, however, do not spend every subsidy dollar due to precautionary savings and bequest motives. Such motives are plausible explanations for why the observed MPC was smaller than 1.
The researchers also traced consumption trends up to 9 weeks after receiving one subsidy payout. Once the individuals received the payments, they increased their spending immediately. By the end of the payout week, recipients had spent more than 0.2 dollars for each dollar of subsidy received. The spike over the first week accounted for nearly 30% of the cumulative spending response, which then gradually tapered off over the course of the remaining 8 weeks.
Agarwal et al. also found substantial heterogeneity in consumption response to the subsidy programme. Lower-liquidity recipients exhibited a substantially larger MPC relative to higher-liquidity individuals. In contrast, the difference in the spending response of lower-income and higher-income individuals was much more muted. These observations suggested that, compared to income, liquidity was a more salient factor influencing MPC.
Robustness and falsification tests
To rule out the possibility that the increase in MPC among subsidy recipients occurred by mere chance and that the subsidy intervention had no effect, Agarwal et al. conducted a bootstrap test on their original sample of 1,340 recipients. Results of the test demonstrated that the likelihood of an MPC of magnitude 0.73 occurring by chance was exceedingly small – which led Agarwal et al. to reject the null hypothesis of randomness.
Using a propensity score matching approach which matched a subsidy recipient to an equivalent (or at least highly similar) non-recipient, the researchers applied a “pseudo subsidy” to the latter to investigate changes in MPC, if any. On the counterfactual that non-recipients had received subsidies, the researchers observed this group’s consumption patterns, both prior to and after actual subsidy disbursements. The change in MPC for non-recipients was miniscule and statistically insignificant. The researchers therefore ruled out the possibility that the observed MPC might be due to life cycle changes that affected the elderly in general, and not just subsidy recipients.
Other robustness tests included varying the time-unit of measurement from weekly to daily and monthly. The average MPC for the daily and monthly samples were qualitatively and quantitatively similar to estimates from the weekly sample.
Policy implications
Findings arising from Agarwal et al.’s study demonstrated that the Silver Support Scheme was working as intended: upon receiving subsidy money, recipients elevated their previously depressed consumption levels, and on average, spent 0.73 dollars for every dollar of subsidy received. Due to precautionary savings and bequest motives, some subsidy money was set aside and not consumed.
Compared to an earlier programme in Singapore, which disbursed vouchers for medical and health insurance expenses, the Silver Support Scheme saw a substantial post-subsidy increase in spending, thereby highlighting that a cash/bank transfer is more effective than a voucher disbursement in stimulating consumption.
Liquidity emerged to be a stronger driver of consumption than income, so if the goal of the policy is to maximise consumption response, then a means test based on liquidity may be more effective than one based on income.
Additionally, the observation that recipients’ expenditure spiked during the first week of receiving subsidies, and then tapered off, suggests that they were not able to smooth their consumption – a short period of elevated consumption was thereafter followed by relative frugality. Agarwal et al. suggested that frequent distribution of smaller benefits, while keeping the overall amount constant, would help recipients consistently maintain a higher level of consumption, and such a disbursement regime was more beneficial to those who live from “payment to payment”.
Authors:
Agarwal, Sumit is the Low Tuck Kwong Professor at the School of Business and Professor in the departments of Economics, Finance and Real Estate at the National University of Singapore.
Qian, Wenlan is Ng Teng Fong Chair Professor in Real Estate and Professor of Finance and Real Estate at the NUS Business School.
Ruan, Tianyue is an Assistant Professor in the Department of Finance at the National University of Singapore (NUS) Business School.
Yeung, Bernard is the Stephen Riady Distinguished Professor in Finance and Strategic Management at the National University of Singapore Business School.
1. The scheme was subsequently enhanced in 2021 to cover a wider segment of the elderly population. For example, the CPF cap has been raised to S$140,000, up from the original $70,000. More details are available here: https://www.silversupport.gov.sg/About/EligibilityCriteria