Monetary Policy According to HANK

The paper notes that “macroeconometric analysis of aggregate time-series data finds a small sensitivity of consumption to changes in the interest rate after controlling for income” (Introduction, citing Campbell and Mankiw 1989; Yogo 2004; Canzoneri, Cumby, and Diba 2007), and offers three microeconomic reasons why the intertemporal substitution channel may be weak even if the individual elasticity of substitution is not small. First, “a sizable fraction of households hold close to zero liquid wealth and face high borrowing costs,” so they “are insensitive to small changes in interest rates,” consistent with Vissing-Jorgensen (2002)’s finding that “non-asset-holders do not react to interest rate cuts.” Second, “standard consumption theory implies that an interest rate cut has negative income effects on the consumption of rich households.” Third, “vast inequality in wealth holdings and composition” means “some households may react to a short-term rate cut by rebalancing their asset portfolio rather than by saving less and consuming more” (Introduction, p. 699).

On the household side the model builds on “the standard Aiyagari-Huggett-İmrohoroğlu incomplete market model, with one important modification: as in Kaplan and Violante (2014), households can save in two assets, a low-return liquid asset and a high-return illiquid asset that is subject to a transaction cost” (Introduction, p. 699). On the supply side, “prices are set by monopolistically competitive producers who face nominal rigidities,” and monetary policy follows a Taylor rule (p. 699). The paper is explicit that this two-asset structure is essential: a one-asset incomplete-markets model calibrated to match total wealth generates “enormous income effects on consumption because all wealth is liquid,” resembling RANK, while calibrated to match only liquid wealth it gets a realistic MPC but “misses over 95 percent of the wealth in the economy,” abstracting from indirect effects operating through investment and the price of capital (p. 700-701).

Decomposing the consumption response to a monetary shock in the baseline calibration, “the indirect components account for 80 percent of the consumption response while the direct component accounts for only 20 percent. This is in stark contrast to typical RANK models” (Sec. IV.B, Table 7, p. 727-728). The paper stresses robustness: “this finding is very robust,” holding up across alternative assumptions about the allocation of profits between liquid and illiquid accounts, the degree of price stickiness, the Taylor rule’s inflation-response coefficient, and the Frisch elasticity of labor supply, and remaining “around 80 percent” even under a comprehensive sensitivity analysis of the borrowing-limit and adjustment-cost parameters governing the “heterogeneous agent block” of the model (Sec. IV.B, p. 728).

The paper attributes the muted direct channel to three interacting forces (Sec. IV.C): poor and wealthy hand-to-mouth households, generated by “uninsurable risk, combined with the coexistence of liquid and illiquid assets,” are “highly sensitive to labor income shocks but are not responsive to interest rate changes”; “the vast inequality in liquid wealth implies that even for non-hand-to-mouth households, a cut in liquid rates leads to strong offsetting income effects on consumption”; and when the spread between liquid and illiquid returns widens after a monetary expansion, “household portfolios adjust away from liquid holdings and toward more lucrative assets rather than toward higher consumption expenditures” (Introduction, p. 699). Together, “all these economic forces counteract the intertemporal substitution effect and lower the direct channel of monetary policy in HANK” (p. 699).

“Since the government is a major issuer of liquid obligations, a change in the interest rate necessarily affects the intertemporal government budget constraint and generates some form of fiscal response that affects household disposable income. Unlike in RANK models, the details of this response matter a great deal for the overall macroeconomic impact of a monetary shock” (Introduction, p. 700). Quantitatively (Sec. IV.D, Table 8), when transfers or taxes adjust to pass the fiscal windfall from lower debt-service costs through to households, indirect effects account for 80 percent of the response as in the baseline; but “when government debt absorbs the slack” instead of transfers, “indirect effects account for a smaller share of the total – 60 percent compared to 80 percent… As a result… the monetary shock has a much smaller impact on the economy, roughly one-half of the baseline value” (p. 734).

The paper shows (Sec. IV.E, Figure 7) that “in one-asset HANK models… there is a well-known tension between matching the high observed aggregate wealth-to-output ratios and generating a large average MPC”: varying the discount rate, “the one-asset model can generate high average wealth or a high MPC, but not both simultaneously.” In the calibrated two-asset model, by contrast, the wealth-to-output ratio is “over 3” while the average quarterly MPC out of a $500 windfall is “0.16” (Sec. III, Table 2), matching both targets jointly – which the authors argue is a prerequisite for a model that can credibly speak to both the size of MPCs (driving the indirect channel) and the scale of aggregate wealth (relevant for investment and asset-price channels).

In RANK, the aggregate Euler equation implies the cumulative consumption elasticity to a rate cut “is independent of the particular path of the real rate. More or less persistent paths with the same cumulative deviation… have the same impact on aggregate consumption” – a neutrality property with “no size-persistence trade-off” (Sec. V.A, p. 736-737). In HANK, this neutrality fails: “when shocks are persistent, a large portion of the interest rate cut, and the associated relaxation of the government budget constraint, occurs in the future. Hence, the hand-to-mouth households receive a smaller increase in transfers upon the impact of the shock and so their consumption response is weaker,” so “a less persistent but larger rate cut can be more effective at expanding aggregate consumption” than a smaller, more persistent one with the same cumulative effect (Introduction, p. 700; Sec. V.A, p. 737).

The paper finds the slope of the inflation-activity trade-off “is not too different in the two economies because it is the common New Keynesian side of the models that largely pins down the relationship” (Sec. V.B, p. 738), so this particular trade-off is less distinctive to HANK than the size-persistence trade-off. However, “in HANK the slope depends on the type of fiscal adjustment: more passive adjustment rules, where government debt absorbs the change in interest payments, are associated with a more favorable trade-off for the monetary authority” (Introduction, p. 700) – so even where the qualitative trade-off resembles RANK, its precise terms remain entangled with fiscal-policy design in a way that has no RANK counterpart.

The paper notes its model shares features with Two-Agent New Keynesian (TANK) models built on the Campbell and Mankiw (1989) spender-saver framework, whose “spenders”… share some similarities with our hand-to-mouth households in that they do not respond to interest rate changes," but stresses a key difference: TANK’s “savers”… engage in intertemporal substitution and are highly responsive to interest rate changes," whereas “in our model even high liquid-wealth households do not increase consumption much in response to an interest rate cut because the risk of receiving negative income shocks and binding liquidity constraints in the future truncates their effective time horizon” (Introduction, p. 701). Consistent with this, Figure 8 shows the TANK cumulative elasticity is “always weakly larger than that in RANK” and depends only modestly on shock persistence, in contrast to HANK’s sharp decline with persistence (Sec. V.A).

They argue that “the relative size of direct versus indirect effects determines the extent to which central banks can precisely target the expansionary impact of their interventions. When direct effects are dominant, as in a RANK model, for the monetary authority to boost aggregate consumption it is sufficient to influence real rates… In a HANK model, instead, the monetary authority must rely on equilibrium feedbacks that boost household income… Reliance on these indirect channels means that the overall effect of monetary policy may be more difficult to fine-tune by manipulating the nominal rate” (Introduction, p. 700). This is especially relevant, they note, when central banks must rely on structural models to extrapolate policy effects beyond well-identified historical experience – as when facing “a binding zero lower bound” and turning to “previously unused policy instruments” (p. 700).

The conclusion notes the model “lacks a distinction between net and gross positions, which would be necessary to assess the role of household leverage on monetary transmission,” and “lacks a distinction between real and nominal assets,” since “all assets in our economy [are] of infinitely short duration” – together meaning the model “cannot generate a commonly observed household portfolio: illiquid housing assets together with long-term nominal mortgage debt” (Sec. VI, p. 738-739), pointing to work such as Garriga, Kydland, and Šustek (2016), Auclert (2019), and Wong (2016) as addressing these balance-sheet channels. The authors also flag that in their model the illiquid asset price “comoves slightly negatively with a monetary shock,” a prediction for which “the empirical evidence… is inconclusive,” identifying model-consistent asset-price dynamics as a priority for the next generation of HANK models (Sec. IV.C).