Internal habits formation and optimality (Q2222223)

The research in this article is related to the well-known model of \textit{C. D. Carroll} et al. [J. Econ. Growth 2, No. 4, 339--367 (1997; Zbl 0926.91038); ``Saving and growth with habit formation'', Am. Econ. Rev. 90, No. 3, 341--355 (2000; \url{doi:10.1257/aer.90.3.341})] for ``Comparison utility in a growth model'' of the type: \[ \max_c \int _0^{\infty}\frac{\left (\frac{c}{h^{\gamma}}\right )^{1-\sigma}}{1-\sigma}e^{-\theta t}dt \] subject to \[ \begin{array}{l} \dot k=Bk-c; \: k(0)=k_0 \\ \dot h=\rho (c-h); \: h(0)=h_0 \end{array} \] where \(k_0>0, h_0>0, k>0, h>0, c>0\), \(B=A-\delta\) and the set of parameters conditions is \[ \Theta=\{(A,\theta,\rho,\delta, \sigma,\gamma): A,\theta >0, \rho\in (0,1),\delta\in [0,1], \sigma\in (1,\infty), \gamma\in [0,1)\}. \] The authors revisit the optimality issue and using advanced techniques in dynamic programming, provide answers to the following long-standing open-ended questions: (i) Is the solution found in Carroll et al. [loc. cit.] optimal? (ii) In addition, is it unique or are there other optimal solutions?