Martingale consumption

Peter Holm Nielsen · 2025

Read Martingale consumption by Peter Holm Nielsen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-20504.txt.

Introduction

Peter Holm Nielsen puts forth a rule for how much to spend in twenty twenty-five. Most past work finds the spend rate by a best fit of joy past risk and time. His way skips tastes and likes. For any fixed share plan for funds at work, he sets the spend rate so that spend to come stays flat on the mean. The spend rate is tweaked to hold that fair mark.

In models where the terms stay fixed, he finds clear forms for the fix. With terms that jump by chance, the fix is the sole one when found, though whether it can be found stays open. When the yield rate jumps by chance, a partial differential equation rules the wealth to spend rate. That law is not straight line and has no closed form fix. A step by step time form gives like marks.

Catalog record

CREATORPeter Holm Nielsen
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-20504

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Martingale consumption Peter Holm Nielsen Abstract We propose martingale consumption as a natural, desirable consumption pattern for any given (proportional) investment strategy. The idea is to always adjust current consumption so as to achieve level expected future consumption under the arbitrarily chosen investment strategy. This approach avoids the formulation of an optimization objective based on preferences towards risk, intertemporal consumption, habit formation etc. We identify general explicit solutions in deterministic-coefficient models. In the general case with random coefficients we establish uniqueness, but the question of existence of a solution is unsettled. With the interest rate as the only random factor we derive a PDE for the wealth-to-consumption factor as a function of the state variables, which, however, is non-linear and without known closed-form solutions. We briefly consider the discrete-time case and obtain similar results. Throughout, we compare with well-known optimal strategies for classical CRRA investors with time-additive utility of consumption and find that under suitable time preferences they may in certain cases achieve martingale consumption simultaneously. Keywords: Consumption strategies; Martingales; PDE’s. 1 Introduction Consumption strategies and patterns constitute a main topic of economics and (mathematical) finance. Often, they are studied along with investment strategies, forming combined investment/consumption problems in various financial market models. The classical economical approach is that of utility optimization under certain assumptions on preferences towards risk, intertemporal substitution etc. This was first carried out in continuous-time models by Merton, (1969) ; Merton, (1971) , who derived explicit optimal investment and consumption strategies for agents with CRRA/HARA risk preferences and time-additive utility of consumption. Ever since, Merton’s pioneering work has been generalized in numerous directions, e.g. in models with stochastic interest rates, e.g. Deelstra et al., (2000) or Munk and Sørensen, (2004) , or general non-markovian financial market models via the martingale method, e.g. Karatzas et al., (1987) or Cox and Huang, (1989) , to name just a few. The optimal consumption strategy derived by Merton in a “standard” model with constant coefficients constitutes a geometric Brownian motion, typically with strictly positive drift. This is perhaps somewhat at odds with a common-sense perspective on preferred consumption patterns for individuals (see below), as well as with observed consumption patterns. Many papers aim to explain observed consumption patterns and/or to formulate various optimization objectives, where the assumption of time-additivity of utility of consumption is replaced by other assumptions so as to allow for, e.g., habit formation, where the current consumption level is measured in relation to past consumption levels in some way, such as in Munk, (2008) , or other economically sensible features. A related branch of papers formulate objectives that aim more generally and perhaps more explicitly at consumption smoothing, where the goal is to obtain a more level and less volatile consumption stream, arguably more in line with preferred consumption patterns, e.g., Bruhn and Steffensen, (2013) or Choi et al., (2022) and references therein. In this paper, we take a different and, to the author’s knowledge, new approach regarding consumption strategies. Taking the (proportional) investment strategy as exogenously and a priori given, we aim to find a martingale consumption strategy, for which we shall use the abbreviation MCS. If the consumption rate process forms a martingale, then the expected future consumption rate at any time, given the evolution until the present time, equals the current consumption rate. In particular, the expected future consumption rates all equal the initial rate at time zero. Thus, on the one hand, a certain kind of consumption smoothing takes place, as the consumption level is continuously adjusted to the current wealth so as to level out remaining expected future consumption, whereas on the other hand, the consumption level continuously moves in line with unanticipated investment gains and losses, fully allowing for paths resulting in non-smooth consumption due to good or bad investment scenarios. The idea is that the agent chooses the (proportional) investment strategy according to his individual preferences, which may (or may not) be inspired by results on optimal investment strategies, and then at any time accepts the current – and expected future – consumption rate that follows from the investment gains or losses, as long as the current rate equals the expected future rates, given the evolution. Our approach is motivated by a certain type of variable annuities employed in practice, i.e., unit-linked “annuity” pension products, where the proportional asset allocation of each retiree follows a predetermined scheme in the retirement phase, and the pension benefits are adjusted concurrently (in pratice, say, annually) to the wealth level resulting from unanticipated investment gains or losses and the updated expectations regarding future returns. This leads to “market based annuities”, where the retirement benefits are allowed to fluctuate with market movements, but with annual readjustments in such a way that the expected remaining benefits are always equal. Just to clarify the idea: It is well known that in a complete market, any consumption strategy satisfying the appropriate budget constraint of the market can be financed through a hedging investment strategy, so in particular, this also goes for any given MCS. However, in this paper we take the (proportional) investment strategy as given and then look for an MCS under that investment strategy. We let the consumption strategy depend on the investment strategy, not the other way around. As alluded to above, the underlying, “implicitly-assumed” desired consumption pattern of the agent is that of an annuity, which in its pure form has deterministic benefits of equal size, and thus in particular constitutes a martingale. Indeed, annuities have traditionally been employed broadly in pension schemes, and generally seem to be widely regarded as sound for retirees, for obvious reasons. However, an annuity imposes a conservative investment strategy in bonds. An MCS can be viewed as a generalization of an annuity that allows for investment strategies in more risky assets with higher expected returns and thus a higher expected consumption rate. Our approach has links to certain aspects of some early research on consumption. In deterministic settings, Modigliani and Brumberg, (1954) and Friedman, (1957) formulated basic, sound principles for consumption versus saving over the life cycle. Their basic idea, also known as the permanent income hypothesis, was that a rational agent should seek to smooth out consumption over the entire life time by taking future income and future consumption needs into account in the decision of current consumption/savings rates, rather than, e.g., let current consumption be driven excessively by the current wage level and/or financial wealth (or changes therein). Along those lines, Hall, (1978) showed that maximization of (expected) accumulated time-additive utility of future consumption in a discrete-time model with a fixed interest rate but, possibly, stochastic future wages, leads to martingale consumption if the agent’s “impatience” parameter exactly offsets the interest rate. Hakansson, (1969) solved the classical combined consumption/investment problem in a discrete-time model and noted that “the optimal consumption strategy satisfies the properties of the permanent (normal) income hypothesis”. This, however, is to be understood in the sense that the optimal consumption strategy is only a function of current wealth and the value of total future income. It also depends parametrically on the agent’s risk aversion and time preferences, and, just as in the continuous-time Merton model, the consumption process is thus not a martingale

Read the complete text