|
Research
A Credit-Banking Explanation of the Equity Premium, Term Premium,
and Risk-Free Rate Puzzles
Micro-founded de-centralized financial intermediation in a cash and costly-credit
model(see Gillman and Kejak, 2008) results in a cost-distortion of returns implying
a lower average nominal and real risk-free rate when compared to standard cash-in-advance
RBC models. Failure of both short-run and long-run Fisher equation
relationships based on observable real and nominal rates and inflation are obtained.
The cost-distortion also leads to an unconditionally upward-sloping average yield
curve of interest rates which is also convex in shape. The model is capable of producing
a positive correlation between the nominal rate and velocity, and a negative
correlation between the ex-post real rate and inflation. More importantly, the model
also predicts a negative correlation between the ex-ante real rate and the ex-ante expected
rate of inflation. Finally, the conditional spread between the usual CCAPM
rate as defined by Canzoneri and Diba (2005) and the model-implied money market
rate is positively correlated with the stance of monetary policy, offering a new
perspective on this systematic link recently studied empirically by Canzoneri et al.
(2007a) and theoretically by Canzoneri and Diba (2005).
Cardiff Economics Working Paper E2008/30 (pdf) ?
Consumption Velocity in a Cash Costly-Credit Model
In a seminal study Hodrick et al. (1991) evaluate the ability of a simple cash-
credit model to produce realistic variability in consumption velocity while at the
same time successfully explaining other key statistics. Sufficient variability in the
latter is found to be associated with far too volatile interest rate behaviour.
Introducing habit-formation in consumption into a production-based cash costly-credit
model (see Gillman and Benk, 2007) makes the evolution of deposits more rigid
relative to credit. The same deposit rigidity leads to a more volatile price of credit,
causing credit production overshooting relative to deposits. But only by introducing
adjustment costs to investment in addition to habit persistence does credit production
overshoot sufficiently to produce realistic variability in consumption velocity.
The model succeeds in capturing sufficient variability in consumption velocity without
obtaining too volatile interest rates. Also, this model of endogenous velocity
does not suffer from indeterminacy problems discussed in Auray et al. (2005). In
contrast to Gillman and Benk (2007), the present study examines the role of the
price-channel of credit production at business cycle frequency, ignoring or holding
fixed the marginal cost channel stemming from credit productivity shocks.
Cardiff Economics Working Paper E2008/31 (pdf) ?
A real monetary business cycle?
Building on work by Jermann (1998) and Boldrin et al. (2001), I include a monetary
sector into a standard RBC model with inelastic physical capital supply and
habit persistence in consumption, which is modeled using a cash-in-advance type
constraint. But in conjunction with cash, the household also self-produces credit
which rises with the nominal rate of interest, so as to result in an LM-type demand
for real money balances (see Kejak and Gillman, 2005; Gillman and Kejak, 2008).
The demand for real money balances rises with consumption, but it also falls with
the nominal rate, as proportionately more credit is produced. As a result of modeling
investment (the supply of physical capital) to exhibit a quadratic adjustment
cost term as in Canzoneri et al. (2007), real and nominal rates are found to fall
following a positive shock to productivity, and the household reallocates it's
liquidity portfolio in favour of money. The model is capable of exhibiting pro-cyclical
endogenous demand for real money balances which also leads consumption, and the
inverted indicator feature of real and nominal rates with respect to both output
and consumption. It also explains procyclical and highly persistent inflation
expectations and in addition provides a possible explanation for the so-called price puzzle
in the structural VAR literature. In general, the model's responses to productivity
innovations are observationally indistinguishable from a New Keynesian model's
response following a monetary expansionary innovation on the Taylor Rule. But
in the present model, real and nominal variables? responses result exclusively from
productivity-driven movements of a flex-price Wicksellian real rate of interest.
Working Paper 06/02/2009 (pdf) ?
Evaluating forecasts of interest rates in linear and non-linear
error-correction frameworks
The following paper's aim is to evaluate the respective forecast performance of
three distinct but related error-correction models. Coefficient estimates and
forecasts are being obtained from a standard linear error-correction, a non-linear
threshold error-correction and a non-linear logistic smooth transition error-correction
model. These are then compared with a simple AR-model in terms of their forecast
accuracy. The algorithms used in estimating the non-linear models are based
on two- and three-dimensional grid-searches over appropriate parameters aimed at
minimising the RSS-function. The algorithms were coded in GAUSS. The variable
to be forecasted is the change in the 3-month U.S. treasury bill and the motivation
for non-linear model specification stems from possible transaction costs incurred in
switching from one financial instrument into another.
MSc Dissertation (pdf) ?
|
|
"The real trouble with this world of ours is not that it is an
unreasonable one, nor even that it is a reasonable one. The commonest
kind of trouble is that it is nearly reasonable, but not quite. Life is
not an illogicality; yet it is a trap for logicians. It looks just a
little more mathematical and regular than it is; its exactitude is
obvious, but its inexactitude is hidden; its wildness lies in wait."
- G.K. Chesterton.
|