You are here

Working Papers

cepar award

Michael Keane and Timothy Neal

Abstract: There is a long standing controversy over the magnitude of the Frisch labor supply elasticity. Macro economists using DSGE models often calibrate it to be large, while many micro data studies find it is small. Several papers attempt to reconcile the micro and macro results. We offer a new and simple explanation: Most micro studies estimate the Frisch using a 2SLS regression of hours changes on income changes. But available instruments are typically “weak.” In that case, we show it is an inherent property of 2SLS that estimates of the Frisch will (spuriously) appear more precise when they are more shifted in the direction of the OLS bias, which is negative. As a result, Frisch elasticities near zero will (spuriously) appear to be precisely estimated, while large estimates will appear to be imprecise. This pattern makes it difficult for a 2SLS t-test to detect a true positive Frisch elasticity. We show how the use of a weak instrument robust hypothesis test, the Anderson-Rubin (AR) test, leads us to conclude the Frisch elasticity is large and significant in the NLSY97 data. In contrast, a conventional 2SLS t-test would lead us to conclude it is not significantly different from zero. Our application illustrates a fundamental problem with 2SLS t-tests that arises quite generally, even with strong instruments. Thus, we argue the AR test should be widely adopted in lieu of the t-test.

Keywords: Frisch elasticity, labor supply, weak instruments, 2SLS, Anderson-Rubin test


Michael Keane and Timothy Neal

Abstract: We provide a simple survey of the literature on weak instruments, aimed at giving practical advice to applied researchers. It is well-known that 2SLS has poor properties if instruments are exogenous but “weak.” We clarify these properties, explain weak instrument tests, and examine how behavior of 2SLS depends on instrument strength. A common standard for “strong” instruments is a first-stage F -statistic of at least 10. But 2SLS has some poor properties in that context: It has low power, and the 2SLS standard error estimate tends to be artificially small in samples where the 2SLS parameter estimate is most contaminated by the OLS bias. This causes t-tests to give very misleading results. Surprisingly, this problem persists even if the first-stage F is in the thousands. Robust tests like Anderson-Rubin greatly alleviate these problems, and should be used in lieu of the t-test even with strong instruments. In many realistic settings a first-stage F well above 10 may be necessary to give high confidence that 2SLS will outperform OLS. For example, in the archetypal application of estimating returns to education, we argue one needs F of at least 50.

Keywords: Instrumental variables, weak instruments, 2SLS, endogeneity, F-test, size distortion, Anderson-Rubin test, conditional t-test, conditional LR test, Fuller, JIVE


Hans Fehr, Maurice Hofmann and George Kudrna 

Abstract: This paper studies the role of pensions and income taxes in determining homeownership and household wealth. It provides a cross-country analysis, using tax and pension policy designs in Germany, the US and Australia. These developed nations have similar incomes per capita but very different homeownership rates, with the US and Australia having much higher homeownership compared to Germany. The question is to what extent the observed differences in homeownership are induced by national tax and transfer policies. To that end, we develop a stochastic, overlapping generations (OLG) model with tenure choice. The model is calibrated to Germany featuring German statutory public pension and dual income tax systems, and then applied to study the effects of alternative income tax and pension policy structures. Our simulation results indicate that the US and Australian policy designs have a dramatic impact on homeownership, explaining more than half of the observed differentials. We also show significant macroeconomic effects due to differences in tax and pension policies.

Keywords: Housing demand, social security, income taxation, stochastic general equilibrium


Boda Kang, Yang Shen, Dan Zhu and Jonathan Ziveyi

Abstract: This paper presents a flexible valuation approach for variable annuity (VA) contracts embedded with guaranteed minimum maturity benefit (GMMB) riders written on an underlying fund that evolves according to a general regime-switching framework. Unlike the classical regime-switching models which only allow model parameters to change upon regime switches, our framework allows, more importantly, model structures to vary. With mild assumptions on the characteristic function of the log-stock price, our model settings enable the study of fundamental features of the market dynamics, such as stochastic volatility and jumps, on the underlying fund value of GMMB in a unified framework. This novel idea is illustrated by a three-regime model whose environments can be characterised by either the geometric Brownian motion process, double exponential process or the Heston (1993) stochastic volatility process. Two versions of the GMMB riders are considered; a fixed or roll-up guarantee and a ratchet geometric average guarantee. With the Fourier Cosine (COS) method which utilises characteristic functions, explicit valuation expressions for various contracts are derived, and numerical illustrations are performed to analyse the efficiency of the approach in terms of computational speed and accuracy. The paper makes a unique contribution by presenting regime-dependent bounds and an algorithm for determining the optimal grid points required for the COS method to achieve a specific level of accuracy. Numerical experiments for the valuation framework reveal that as the likelihood of regime shifts increases, the price difference of VA contracts with different initial regimes diminishes, which is consistent with financial intuition.

Keywords: Variable annuity contracts; GMMB; COS method; Generalised regime-switching model; Ratch-et options



Hazel Bateman, Isabella Dobrescu, Junhao Liu, Ben R. Newell & Susan Thorp

Abstract: The Australian COVID-19 Early Release Scheme (ERS) allowed people in financial hardship immediate access to up to $A20,000 of their ‘preserved’ retirement savings between April and December 2020. Using data from a large Australian pension fund, we examine what drives people’s decisions to take advantage of the ERS. We find that while the majority of survey respondents withdrew money for immediate consumption needs, a substantial proportion of them were concerned about future needs. Most withdrawers thought about the decision for less than a week and many appeared to use the $A10,000 per round limit as an anchor in choosing their withdrawal amount. Conditional on eligibility, the probability of withdrawal was significantly higher where respondents (i) were more concerned about future needs, (ii) did not think about the long-term impact, and (iii) under-estimated or did not estimate the fall in their retirement savings. Our results suggest that many people who withdrew under the scheme did not fully understand the consequences of their choice. These findings raise the question of whether the framing of ‘mandatory’ retirement savings as a mental account to finance retirement has been irrevocably damaged.

Keywords: pension early access, retirement savings, mental accounts, COVID-19

Hazel Bateman

Julie Agnew, Hazel Bateman, Christine Eckert, Fedor Iskhakov, Jordan Louviere, Susan Thorp

Abstract: What kinds of people will pay bad financial advisers? We show that experimental participants (n=2003) with a proclivity toward confirmation bias are more susceptible to bad advisers. We give participants a sequence of signals of adviser quality that can be clear or ambiguous, depending on each participant’s ability to discern bad advice. Rational participants set aside ambiguous signals and do not use them to update beliefs about advisers. Biased participants treat ambiguous signals as favoring their priors, and update accordingly. Younger, more trusting, more impulsive, less financially literate and less numerate participants are most vulnerable to paying a poor-quality adviser.


Mike Sherris CEPAR

Michael Sherris 

Abstract: The Final Report of the Royal Commission into Aged Care Quality and Safety (2021) highlighted the challenges in developing a sustainable financing system for Aged Care in Australia. The Report recommended additional funding both in the short term and longer term, to provide an adequate level of aged care quality for older Australians including exploring an actuarially based contributory social insurance scheme for aged care. Sustainable financing of aged care requires a balance between government tax-based financing, individual contributions during working life through an aged care levy, co-payments for aged care costs for those receiving aged care and means testing for these co-payments. There should be a role for private market insurance and financing to supplement government financed aged care support.


Yuanyuan Deng, Hanming Fang, Katja Hanewald and Shang Wu

Abstract: We develop and calibrate a life-cycle model of labor supply and consumption to quantify the implications of alternative pension reforms on labor supply, individual welfare, and government budget for China’s basic old-age insurance program. We focus on urban males and distinguish low-skilled and high-skilled individuals, who differ in their preferences, health and labor income dynamics, and medical expense processes. We use the calibrated model to evaluate three potential pension reforms: (i) increasing the pension eligibility age from 60 to 65, but keeping the current pension benefit rule unchanged; (ii) keeping the pension eligibility age at 60, but proportionally lowering pension benefits so that the pension program’s budget is the same as under Reform (i); and (iii) increasing the pension eligibility age to 65 and simultaneously increasing the pension benefits so that individuals of both skill types attain the same individual welfare levels as in the status quo. We find that relative to the baseline, both Reforms (i) and (ii) can substantially improve the budgets of the pension system, but at the cost of substantial individual welfare loss for both skill types. In contrast, we find that Reform (iii) can modestly improve the budget of the pension system while ensuring that both skill types are as well off as in the status quo. We find that Reforms (i) and (ii) slightly increases, but Reform (iii) slightly decreases, the overall labor supply.

Keywords: pension reform; labor force participation; welfare; life-cycle behavior; China



Hazel Bateman, Paul Gerrans, Susan Thorp and Yunbo Zeng

Abstract: We administered an online survey to elicit consumers’ subjective assessments of their decision state for the purchase of life insurance - from pre-aware to purchase decision - in a setting of both active choice and default cover. We find that household formation and financial assets are associated with higher decision states, but not always with being capable and ready to choose. The financially literate are more likely to be in a higher state, but the less financially literate are spread across several states. We also find that personal values matter for readiness to make a choice about life insurance with respondents who place more value on benevolence and self- determination more likely to be aware of life insurance and capable to choose. We conclude that personal values help consumers choose suitable cover and that interventions to increase cover and improve suitability of life insurance should target progression through the decision states.

Keywords: Life insurance, decision states, personal values, financial literacy, defaults.