One option for the ECB the previous blog post did not explicitly
consider is what (as of today) can be called the Swiss option: raising the
deposit rate, but not applying it to all the excess reserves, but only fraction
of excess reserves. This is exactly what the Swiss National Bank has done
today: Banks’ deposits held at the SNB will be renumerated at the policy rate
up to certain threshold, and by 0% above this threshold. Effectively, this is
the reverse of the option discussed in previous post, where only excess
reserves above, but not below, some threshold are renumerated.
Why did I not consider this option? Because it should not
work. In principle, when the system has large amount of excess reserves - higher
than the amount of sum of all the “allowances” – then this renumeration scheme
should negate the increase in policy rates, and hence be effectively identical
to the option of not raising deposit rate at all, which I have discussed.
To see this, consider situation of two commercial banks, one
with excess reserves above the threshold, and one with excess reserves below. The
one with excess reserves above threshold is earning zero on its reserves above
this threshold, so should be willing to lend them out at any interest rate above
0%.[1]
Meanwhile, the bank with excess reserves below threshold should be willing to
borrow for any rate below the deposit rate, since it can earn deposit rate. As
price of lending is the interbank interest rate, this should push the interbank
rate close to the deposit rate. That is, as long as there is a bank which did
not exceed its threshold, so that it can earn the deposit rate on marginal
excess reserves. But this last condition is crucial: If the amount of
excess reserves is so high that all banks are above their thresholds, then
there will not be any bank willing to borrow at rates close to the deposit rate,
because no bank will be earning the deposit rate on its marginal excess
reserves, all banks will be earning 0%. This will push the interbank rate
towards the 0%, negating the increase in deposit rates. This is the same
mechanism that allowed central banks to push interbank rates to negative values:
when there are excess reserves, interbank rates tend to deposit rate.
Arguably, the SNB and
ECB are in this latter situation. Of course, one could address this by
increasing the threshold. But this would negate the desired goal of this policy
which is decreasing the sums paid to commercial banks. So, in my opinion, the
option cannot work, at least in absence of some other action. SNB mentions liquidity
absorbing operations, which could be this other action. But for them to work,
they will have to pay interests higher than 0%, which will again clash with the
desired goal.
The overall conclusion? Central banks do not have neat
options of avoiding paying large sums to commercial banks, all the options are fraught
with complications. This should not be surprising: back when the goal was to lower interest
rates, excess reserves were a useful tool, as they allowed the central bank to
push interbank rates close to deposit rate. Now, however, central banks are
increasing rates. In such situation, excess reserves are an obstacle. Not
insurmountable obstacle – you can just pay higher deposit rate – but when
combined with political desire to avoid paying money to commercial banks, it becomes
troublesome obstacle indeed.
No comments:
Post a Comment