The Principles Which Should Regulate the Amount of the Banking
Reserve to Be Kept by the Bank of England.


There is a very common notion that the amount of the reserve which
the Bank of England ought to keep can be determined at once from the
face of their weekly balance sheet. It is imagined that you have
only to take the liabilities of the Banking department, and that a
third or some other fixed proportion will in all cases be the amount
of reserve which the Bank should keep against those liabilities. But
to this there are several objections, some arising from the general
nature of the banking trade, and others from the special position of
the Bank of England.

That the amount of the liabilities of a bank is a principal element
in determining the proper amount of its reserve is plainly true; but
that it is the only element by which that amount is determined is
plainly false. The intrinsic nature of these liabilities must be
considered, as well as their numerical quantity. For example, no one
would say that the same amount of reserve ought to be kept against
acceptances which cannot be paid except at a certain day, and
against deposits at call, which may be demanded at any moment. If a
bank groups these liabilities together in the balance-sheet, you
cannot tell the amount of reserve it ought to keep. The necessary
information is not given you.

Nor can you certainly determine the amount of reserve necessary to
be kept against deposits unless you know something as to the nature
of these deposits. If out of 3,000,000 L. of money, one depositor
has 1,000,000 L. to his credit, and may draw it out when he pleases,
a much larger reserve will be necessary against that liability of
1,000,000 L. than against the remaining 2,000,000 L. The intensity of
the liability, so to say, is much greater; and therefore the
provision in store must be much greater also. On the other hand,
supposing that this single depositor is one of calculable
habitssuppose that it is a public body, the time of whose demands is
known, and the time of whose receipts is known alsothis single
liability requires a less reserve than that of an equal amount of
ordinary liabilities. The danger that it win be called for is much
less; and therefore the security taken against it may be much less
too. Unless the quality of the liabilities is considered as well as
their quantity, the due provision for their payment cannot be
determined.

These are general truths as to all banks, and they have a very
particular application to the Bank of England. The first application
is favourable to the Bank; for it shows the danger of one of the
principal liabilities to be much smaller than it seems. The largest
account at the Bank of England is that of the English Government;
and probably there has never been any account of which it was so
easy in time of peace to calculate the course. All the material
facts relative to the English revenue, and the English expenditure,
are exceedingly well known; and the amount of the coming payments to
and from this account are always, except in war times, to be
calculated with wonderful accuracy. In war, no doubt, this is all
reversed; the account of a government at war is probably the most
uncertain of all accounts, especially of a government of a scattered
empire, like the English, whose places of outlay in time of war are
so many and so distant, and the amount of whose payments is
therefore so incalculable. Ordinarily, however, there is no account
of which the course can be so easily predicted; and therefore no
account which needs in ordinary times so little reserve. The
principal payments, when they are made, are also of the most
satisfactory kind to a banker; they are, to a great extent, made to
another account at his bank. These largest ordinary payments of the
Government are the dividends on the debt, and these are mostly made
to bankers who act as agents for the creditors of the nation. The
payment of the dividends for the Government is, therefore, in great
part a transfer from the account of the Government to the accounts
of the various bankers. A certain amount no doubt goes almost at
once to the non-banking classes; to those who keep coin and notes in
house, and have no account at any bank. But even this amount is
calculable, for it is always nearly the same. And the entire
operation is, to those who can watch it, singularly invariable time
after time.

But it is important to observe, that the published accounts of the
Bank give no such information to the public as win enable them to
make their own calculations. The account of which we have been
speaking is the yearly account of the English Governmentwhat we may
call the Budget account, that of revenue and expenditure. And the
laws of this are, as we have shown, already known. But under the
head 'Public Deposits' in the accounts of the Bank, are contained
also other accounts, and particularly that of the Secretary for
India in Council, the laws of which must be different and are quite
unknown. The Secretary for India is a large lender on its account.
If any one proposed to give such power to the Chancellor of the
Exchequer, there would be great fear and outcry. But so much depends
on habit and tradition, that the India Office on one side of Downing
Street can do without remark, and with universal assent, what it
would be thought 'unsound' and extravagant to propose that the other
side should do. The present India Office inherits this independence
from the old Board of the Company, which, being mercantile and
business-like, used to lend its own money on the Stock Exchange as
it pleased; the Council of India, its successor, retains the power.
Nothing can be better than that it should be allowed to do as it
likes; but the mixing up the account of a body which has such a
power, and which draws money from India, with that of the Home
government clearly prevents the general public from being able to
draw inferences as to the course of the combined account from its
knowledge of home finance only. The account of 'public deposits' in
the Bank return includes other accounts too, as the Savings' Bank
balance, the Chancery Funds account, and others; and in consequence,
till lately the public had but little knowledge of the real changes
of the account of our Government, properly so called. But Mr. Lowe
has lately given us a weekly account, and from this, and not from
the Bank account, we are able to form a judgment. This account and
the return of the Bank of England, it is true, unhappily appear on
different days; but except for that accident our knowledge would be
perfect; and as it is, for almost all purposes what we know is
reasonably sufficient. We can now calculate the course of the
Government account nearly as well as it is possible to calculate it.

So far, as we have said, an analysis of the return of the Bank of
England is very favourable to the Bank. So great a reserve need not
usually be kept against the Government account as if it were a
common account. We know the laws of its changes peculiarly well: we
can tell when its principal changes will happen with great accuracy;
and we know that at such changes most of what is paid away by the
Government is only paid to other depositors at the Bank, and that it
win really stay at the Bank, though under another name. If we look
to the private deposits of the Bank of England, at first sight we
may think that the result is the same. By far the most important of
these are the 'Bankers' deposits'; and, for the most part, these
deposits as a whole are likely to vary very little. Each banker, we
will suppose, keeps as little as he can, but in all domestic
transactions payment from one is really payment to the other. All
the most important transactions in the country are settled by
cheques; these cheques are paid in to the 'clearing-house,' and the
balances resulting from them are settled by transfers from the
account of one banker to another at the Bank of England. Payments
out of the bankers' balances, therefore, correspond with payments
in. As a whole, the deposit of the bankers' balances at the Bank of
England would at first sight seem to be a deposit singularly stable.

Indeed, they would seem, so to say, to be better than stable. They
augment when everything else tends to diminish. At a panic, when all
other deposits are likely to be taken away, the bankers' deposits,
augment; in fact they did so in 1866, though we do not know the
particulars; and it is natural that they should so increase. At such
moments all bankers are extremely anxious, and they try to
strengthen themselves by every means in their power; they try to
have as much money as it is possible at command; they augment their
reserve as much as they can, and they place that reserve at the Bank
of England. A deposit which is not likely to vary in ordinary times,
and which is likely to augment in times of danger, seems, in some
sort, the model of a deposit. It might seem not only that a large
proportion of it might be lent, but that the whole of it might be
so. But a further analysis will, as I believe, show that this
conclusion is entirely false; that the bankers' deposits are a
singularly treacherous form of liability; that the utmost caution
ought to be used in dealing with them; that, as a rule, a less
proportion of them ought to be lent than of ordinary deposits.

The easiest mode of explaining anything is, usually, to exemplify it
by a single actual case. And in this subject, fortunately, there is
a most conspicuous case near at hand. The German Government has
lately taken large sums in bullion from this country, in part from
the Bank of England, and in part not, according as it chose. It was
in the main well advised, and considerate in its action; and did not
take nearly as much from the Bank as it might, or as would have been
dangerous. Still it took large sums from the Bank; and it might
easily have taken more. How then did the German Government obtain
this vast power over the Bank? The answer is, that it obtained it by
means of the bankers' balances, and that it did so in two ways.

First, the German Government had a large balance of its own lying at
a particular Joint Stock Bank. That bank lent this balance at its
own discretion, to bill-brokers or others, and it formed a single
item in the general funds of the London market. There was nothing
special about it, except that it belonged to a foreign government,
and that its owner was always likely to call it in, and sometimes
did so. As long as it stayed unlent in the London Joint Stock Bank,
it increased the balances of that bank at the Bank of England; but
so soon as it was lent, say, to a bill-broker, it increased the
bill-broker's balance; and as soon as it was employed by the
bill-broker in the discount of bills, the owners of those bills paid
it to their credit at their separate banks, and it augmented the
balances of those bankers at the Bank of England. Of course if it
were employed in the discount of bills belonging to foreigners, the
money might be taken abroad, and by similar operations it might also
be transferred to the English provinces or to Scotland. But, as a
rule, such money when deposited in London, for a considerable time
remains in London; and so long as it does so, it swells the
aggregate balances of the body of bankers at the Bank of England. It
is now in the balance of one bank, now of another, but it is always
dispersed about those balances somewhere. The evident consequence is
that this part of the bankers' balances is at the mercy of the
German Government when it chooses to apply for it. Supposing, then,
the sum to be three or four millions and I believe that on more than
one occasion in the last year or two it has been quite as much, if
not more--that sum might at once be withdrawn from the Bank of
England. In this case the Bank of England is in the position of a
banker who is liable for a large amount to a single customer, but
with this addition, that it is liable for an unknown amount. The
German Government, as is well known, keeps its account (and a very
valuable one it must be) at the London Joint Stock Bank; but the
Bank of England has no access to the account of the German
Government at that bank; they cannot tell how much German money is
lying to the credit there. Nor can the Bank of England infer much
from the balance of the London Joint Stock Bank in their Bank, for
the German money was probably paid in various sums to that bank, and
lent out again in other various sums. It might to some extent
augment that bank's balance at the Bank of England, or it might not,
but it certainly would not be so much added to that balance; and
inspection of that bank's balance would not enable the Bank of
England to determine even in the vaguest manner what the entire sum
was for which it might be asked at any moment. Nor would the
inspection of the bankers' balances as a whole lead to any certain
and sure conclusions. Something might be inferred from them, but not
anything certain. Those balances are no doubt in a state of constant
fluctuation; and very possibly during the time that the German money
was coming in some other might be going out. Any sudden increase in
the bankers' balances would be a probable indication of new foreign
money, but new foreign money might come in without causing an
increase, since some other and contemporaneous cause might effect a
counteracting decrease.

This is the first, and the plainest way in which the German
Government could take, and did take, money from this country; and in
which it might have broken the Bank of England if it had liked. The
German Government had money here and took it away, which is very
easy to understand. But the Government also possessed a far greater
power, of a somewhat more complex kind. It was the owner of many
debts from England. A large part of the 'indemnity' was paid by
France to Germany in bills on England, and the German Government, as
those bills became due, acquired an unprecedented command over the
market. As each bill arrived at maturity, the German Government
could, if it chose, take the proceeds abroad; and it could do so in
bullion, as for coinage purposes it wanted bullion. This would at
first naturally cause a reduction in the bankers' balances; at least
that would be its tendency. Supposing the German Government to hold
bill A, a good bill, the banker at whose bank bill A was payable
would have to pay it; and that would reduce his balance; and as the
sum so paid would go to Germany, it would not appear to the credit
of any other banker: the aggregate of the bankers' balances would
thus be reduced. But this reduction would not be permanent. A banker
who has to pay 100,000 L. cannot afford to reduce his balance at the
Bank of England 100,000 L.; suppose that his liabilities are
2,000,000 L., and that as a rule he finds it necessary to keep at
the Bank one-tenth of these liabilities, or 200,000 L., the payment
of 100,000 L. would reduce his reserve to 100,000 L.; but his
liabilities would be still 1,900,000 L. and therefore to keep up his
tenth he would have 90,000 L. to find. His process for finding it is
this: he calls in, say, a loan to the bill-brokers; and if no equal
additional money is contemporaneously carried to these brokers
(which in the case of a large withdrawal of foreign money is not
probable), they must reduce their business and discount less. But
the effect of this is to throw additional business on the Bank of
England. They hold the ultimate reserve of the country, and they
must discount out of it if no one else will: if they declined to do
so there would be panic and collapse. As soon, therefore, as the
withdrawal of the German money reduces the bankers' balances, there
is a new demand on the Bank for fresh discounts to make up those
balances. The drain on the Bank is twofold: first, the banking
reserve is reduced by exportation of the German money, which reduces
the means of the Bank of England; and then out of those reduced
means the Bank of England has to make greater advances.

The same result may be arrived at more easily. Supposing any foreign
Government or person to have any sort of securities which he can
pledge in the market, that operation gives it, or him, a credit on
some banker, and enables it, or him, to take money from the banking
reserve at the Bank of England, and from the bankers' balances; and
to replace the bankers' balances at their inevitable minimum, the
Bank of England must lend. Every sudden demand on the country
causes, in proportion to its magnitude, this peculiar effect. And
this is the reason why the Bank of England ought, I think, to deal
most cautiously and delicately with their banking deposits. They are
the symbol of an indefinite liability: by means of them, as we see,
an amount of money so great that it is impossible to assign a limit
to it might be abstracted from the Bank of England. As the Bank of
England lends money to keep up the bankers' balances, at their usual
amount, and as by means of that usual amount whatever sum foreigners
can get credit for may be taken from us, it is not possible to
assign a superior limit (to use the scientific word) to the demands
which by means of the bankers' balances may be made upon the Bank of
England.

The result comes round to the simple point, on which this book is a
commentary: the Bank of England, by the effect of a long history,
holds the ultimate cash reserve of the country; whatever cash the
country has to pay comes out of that reserve, and therefore the Bank
of England has to pay it. And it is as the Bankers' Bank that the
Bank of England has to pay it, for it is by being so that it becomes
the keeper of the final cash reserve.

Some persons have been so much impressed with such considerations as
these, that they have contended that the Bank of England ought never
to lend the 'bankers' balances' at all, that they ought to keep them
intact, and as an unused deposit. I am not sure, indeed, that I have
seen that extreme form of the opinion in print, but I have often
heard it in Lombard Street, from persons very influential and very
qualified to judge; even in print I have seen close approximations
to it. But I am satisfied that the laying down such a 'hard and
fast' rule would be very dangerous; in very important and very
changeable business rigid rules are apt to be often dangerous. In a
panic, as has been said, the bankers' balances greatly augment. It
is true the Bank of England has to lend the money by which they are
filled. The banker calls in his money from the bill-broker, ceases
to re-discount for that broker, or borrows on securities, or sells
securities; and in one or other of these ways he causes a new demand
for money which can only at such times be met from the Bank of
England. Every one else is in want too. But without inquiring into
the origin of the increase at panics, the amount of the bankers'
deposits in fact increases very rapidly; an immense amount of unused
money is at such moments often poured by them into the Bank of
England. And nothing can more surely aggravate the panic than to
forbid the Bank of England to lend that money. Just when money is
most scarce you happen to have an unusually large fund of this
particular species of money, and you should lend it as fast as you
can at such moments, for it is ready lending which cures panics, and
non-lending or niggardly lending which aggravates them.

At other times, particularly at the quarterly payment of the
dividends, an absolute rule which laid down that the bankers'
balances were never to be lent, would be productive of great
inconvenience. A large sum is just then paid from the Government
balance to the bankers' balances, and if you permitted the Bank to
lend it while it was still in the hands of the Government, but
forbad them to lend it when it came into the hands of the bankers, a
great tilt upwards in the value of money would be the consequence,
for a most important amount of it would suddenly have become
ineffective.

But the idea that the bankers' balances ought never to be lent is
only a natural aggravation of the truth that these balances ought to
be used with extreme caution; that as they entail a liability
peculiarly great and singularly difficult to foresee, they ought
never to be used like a common deposit.

It follows from what has been said that there are always possible
and very heavy demands on the Bank of England which are not shown in
the account of the Banking department at all: these demands may be
greatest when the liabilities shown by that account are smallest,
and lowest when those liabilities are largest. If, for example, the
German Government brings bills or other good securities to this
market, obtains money with them, and removes that money from the
market in bullion, that money may, if the German Government choose,
be taken wholly from the Bank of England. If the wants of the German
Government be urgent, and if the amount of gold 'arrivals,' that is,
the gold coming here from the mining countries, be but small, that
gold will be taken from the Bank of England, for there is no other
large store in the country. The German Government is only a
conspicuous example of a foreign power which happens lately to have
had an unusual command of good securities, and an unusually
continuous wish to use them in England. Any foreign state hereafter
which wants cash will be likely to come here for it; so long as the
Bank of France should continue not to pay in specie, a foreign state
which wants it must of necessity come to London for it.

And no indication of the likelihood or unlikelihood of that want can
be found in the books of the Bank of England.

What is almost a revolution in the policy of the Bank of England
necessarily follows: no certain or fixed proportion of its
liabilities can in the present times be laid down as that which the
Bank ought to keep in reserve. The old notion that one-third, or any
other such fraction, is in all cases enough, must be abandoned. The
probable demands upon the Bank are so various in amount, and so
little disclosed by the figures of the account, that no simple and
easy calculation is a sufficient guide. A definite proportion of the
liabilities might often be too small for the reserve, and sometimes
too great. The forces of the enemy being variable, those of the
defence cannot always be the same.

I admit that this conclusion is very inconvenient. In past times it
has been a great aid to the Bank and to the public to be able to
decide on the proper policy of the Bank from a mere inspection of
its account. In that way the Bank knew easily what to do and the
public knew easily what to foresee. But, unhappily, the rule which
is most simple is not always the rule which is most to be relied
upon. The practical difficulties of life often cannot be met by very
simple rules; those dangers being complex and many, the rules for
encountering them cannot well be single or simple. A uniform remedy
for many diseases often ends by killing the patient.

Another simple rule often laid down for the management of the Bank
of England must now be abandoned also. It has been said that the
Bank of England should look to the market rate, and make its own
rate conform to that. This rule was, indeed, always erroneous. The
first duty of the Bank of England was to protect the ultimate cash
of the country, and to raise the rate of interest so as to protect
it. But this rule was never so erroneous as now, because the number
of sudden demands upon that reserve was never formerly so great. The
market rate of Lombard Street is not influenced by those demands.
That rate is determined by the amount of deposits in the hands of
bill-brokers and bankers, and the amount of good bills and
acceptable securities offered at the moment. The probable efflux of
bullion from the Bank scarcely affects it at all; even the real
efflux affects it but little; if the open market did not believe
that the Bank rate would be altered in consequence of such effluxes
the market rate would not rise. If the Bank choose to let its
bullion go unheeded, and is seen to be going so to choose, the value
of money in Lombard Street will remain unaltered. The more numerous
the demands on the Bank for bullion, and the more variable their
magnitude, the more dangerous is the rule that the Bank rate of
discount should conform to the market rate. In former quiet times
the influence, or the partial influence, of that rule has often
produced grave disasters. In the present difficult times an
adherence to it is a recipe for making a large number of panics.

A more distinct view of abstract principle must be taken before we
can fix on the amount of the reserve which the Bank of England ought
to keep. Why should a bank keep any reserve? Because it may be
called on to pay certain liabilities at once and in a moment. Why
does any bank publish an account? In order to satisfy the public
that it possesses cashor available securitiesenough to meet its
liabilities. The object of publishing the account of the banking
department of the Bank of England is to let the nation see how the
national reserve of cash stands, to assure the public that there is
enough and more than enough to meet not only all probable calls, but
all calls of which there can be a chance of reasonable apprehension.
And there is no doubt that the publication of the Bank account gives
more stability to the money market than any other kind of precaution
would give. Some persons, indeed, feared that the opposite result
would happen; they feared that the constant publication of the
incessant changes in the reserve would terrify and harass the public
mind. An old banker once told me: 'Sir, I was on Lord Althorp's
committee which decided on the publication of the Bank account, and
I voted against it. I thought it would frighten people. But I am
bound to own that the committee was right and I was wrong, for that
publication has given the money market a greater sense of security
than anything else which has happened in my time.' The diffusion of
confidence through Lombard Street and the world is the object of the
publication of the Bank accounts and of the Bank reserve.

But that object is not attained if the amount of that reserve when
so published is not enough to tranquillise people. A panic is sure
to be caused if that reserve is, from whatever cause, exceedingly
low. At every moment there is a certain minimum which I will call
the apprehension minimum,' below which the reserve cannot fall
without great risk of diffused fear; and by this I do not mean
absolute panic, but only a vague fright and timorousness which
spreads itself instantly, and as if by magic, over the public mind.
Such seasons of incipient alarm are exceedingly dangerous, because
they beget the calamities they dread. What is most feared at such
moments of susceptibility is the destruction of credit; and if any
grave failure or bad event happens at such moments, the public fancy
seizes on it, there is a general run, and credit is suspended. The
Bank reserve then never ought to be diminished below the
'apprehension point.' And this is as much as to say, that it never
ought very closely to approach that point; since, if it gets very
near, some accident may easily bring it down to that point and cause
the evil that is feared.

There is no 'royal road' to the amount of the 'apprehension
minimum': no abstract argument, and no mathematical computation will
teach it to us. And we cannot expect that they should. Credit is an
opinion generated by circumstances and varying with those
circumstances. The state of credit at any particular time is a
matter of fact only to be ascertained like other matters of fact; it
can only be known by trial and inquiry. And in the same way, nothing
but experience can tell us what amount of 'reserve' will create a
diffused confidence; on such a subject there is no way of arriving
at a just conclusion except by incessantly watching the public mind,
and seeing at each juncture how it is affected.

Of course in such a matter the cardinal rule to be observed is, that
errors of excess are innocuous but errors of defect are destructive.
Too much reserve only means a small loss of profit, but too small a
reserve may mean 'ruin.' Credit may be at once shaken, and if some
terrifying accident happen to supervene, there may be a run on the
Banking department that may be too much for it, as in 1857 and 1866,
and may make it unable to pay its way without assistanceas it was m
those years.

And the observance of this maxim is the more necessary because the
'apprehension minimum' is not always the same. On the contrary, in
times when the public has recently seen the Bank of England exposed
to remarkable demands, it is likely to expect that such demands may
come again. Conspicuous and recent events educate it, so to speak;
it expects that much will be demanded when much has of late often
been demanded, and that little will be so, when in general but
little has been so. A bank like the Bank of England must always,
therefore, be on the watch for a rise, if I may so express it, in
the apprehension minimum; it must provide an adequate fund not only
to allay the misgivings of to-day, but also to allay what may be the
still greater misgivings of to-morrow. And the only practical mode
of obtaining this object is--to keep the actual reserve always in
advance of the minimum 'apprehension' reserve.

And this involves something much more. As the actual reserve is
never to be less, and is always, if possible, to exceed by a
reasonable amount the 'minimum' apprehension reserve, it must when
the Bank is quiet and taking no precautions very considerably exceed
that minimum. All the precautions of the Bank take time to operate.
The principal precaution is a rise in the rate of discount, and such
a rise certainly does attract money from the Continent and from all
the world much faster than could have been anticipated. But it does
not act instantaneously; even the right rate, the ultimately
attractive rate, requires an interval for its action, and before the
money can come here. And the right rate is often not discovered for
some time. It requires several 'moves,' as the phrase goes, several
augmentations of the rate of discount by the Bank, before the really
effectual rate is reached, and in the mean time bullion is ebbing
away and the 'reserve' is diminishing. Unless, therefore, in times
without precaution the actual reserve exceed the 'apprehension
minimum' by at least the amount which may be taken away in the
inevitable interval, and before the available precautions begin to
operate, the rule prescribed will be infringed, and the actual
reserve will be less than the 'apprehension' minimum. In time the
precautions taken may attract gold and raise the reserve to the
needful amount, but in the interim the evils may happen against
which the rule was devised, diffused apprehension may arise, and
then any unlucky accident may cause many calamities.

I may be asked, 'What does all this reasoning in practice come to?
At the present moment how much reserve do you say the Bank of
England should keep? state your recommendation clearly (I know it
will be said) if you wish to have it attended to.' And I will answer
the question plainly, though in so doing there is a great risk that
the principles I advocate may be in some degree injured through some
mistake I may make in applying them.

I should say that at the present time the mind of the monetary world
would become feverish and fearful if the reserve in the Banking
department of the Bank of England went below 10,000,000 L. Estimated
by the idea of old times, by the idea even of ten years ago, that
sum, I know, sounds extremely large. My own nerves were educated to
smaller figures, because I was trained in times when the demands on
us were less, when neither was so much reserve wanted nor did the
public expect so much. But I judge from such observations as I can
make of the present state of men's minds, that in fact, and whether
justifiably or not, the important and intelligent part of the public
which watches the Bank reserve becomes anxious and dissatisfied if
that reserve falls below 10,000,000 L. That sum, therefore, I call
the 'apprehension minimum' for the present times. Circumstances may
change and may make it less or more, but according to the most
careful estimate I can make, that is what I should call it now.

It will be said that this estimate is arbitrary and these figures
are conjectures. I reply that I only submit them for the judgment of
others. The main question is one of fact--Does not the public mind
begin to be anxious and timorous just where I have placed the
apprehension point? and the deductions from that are comparatively
simple questions of mixed fact and reasoning. The final appeal in
such cases necessarily is to those who are conversant with and who
closely watch the facts.

I shall perhaps be told also that a body like the Court of the
Directors of the Bank of England cannot act on estimates like these:
that such a body must have a plain rule and keep to it. I say in
reply, that if the correct framing of such estimates is necessary
for the good guidance of the Bank, we must make a governing body
which can correctly frame such estimates. We must not suffer from a
dangerous policy because we have inherited an imperfect form of
administration. I have before explained in what manner the
government of the Bank of England should, I consider, be
strengthened, and that government so strengthened would, I believe,
be altogether competent to a wise policy.

Then I should say, putting the foregoing reasoning into figures,
that the Bank ought never to keep less than 11,000,000 L.. or
11,500,000 L. since experience shows that a million, or a million
and a half, may be taken from us at any time. I should regard this
as the practical minimum at which, roughly of course, the Bank
should aim, and which it should try never to be below. And, in order
not to be below 11,500,000 L., the Bank must begin to take
precautions when the reserve is between 14,000,000 L. and 15,000,000
l.; for experience shows that between 2,000,000 L. and 3,000,000 L.
may, probably enough, be withdrawn from the Bank store before the
right rate of interest is found which will attract money from
abroad, and before that rate has had time to attract it. When the
reserve is between 14,000,000 L. and 15,000,000 L., and when it
begins to be diminished by foreign demand, the Bank of England
should, I think, begin to act, and to raise the rate of interest.