-.-

*The following brief primer is available for viewing in PDF format by clicking the link here: A Primer on the BoC Interest Rate Corridor System*

The monetary policy of the Bank of Canada (the **“BoC”**) aims to maintain the following:

- Stable inflation rate;
- Sound financial system;
- Safe and secure currency; and
- Efficient management of government funds and public debt.

Central banks implement monetary policy via the control of overnight interest rates. As of December 2015, the monetary policy framework of the BoC is a pure interest-rate corridor system (**“Corridor System”**). This Corridor System is also synonymously known as a “tunnel” or “channel” system, as this interest-rate framework utilizes standing facilities to control the domestic overnight interest rate and thereby establishing an operating band which contains the movements of the overnight interest rate.

The Corridor System requires an operating band: an interest-rate ceiling and floor. The ceiling or upper limit of the operating band is the interest rate the BoC’s standing lending facility (SLF) charges participants on overnight overdraft loans (the **“Bank Rate”**). If a participant of BoC’s Large Value Transfer System (LVTS) is short of overnight funds, it can thus borrow the necessary amount from the BoC at the Bank Rate. Consequently, the floor or lower limit of the operating band is the interest rate the BoC pays participants on their deposits held overnight at the BoC (the **“Deposit Rate”**). This means that if a LVTS participant has excess funds at the end of the day, it can earn the Deposit Rate on such funds overnight.

Combination of the Bank Rate and Deposit Rate, which the BoC sets, necessitates the condition for the target overnight interest rate to stay within the operating band.

Figure A (below) is a basic illustration of how the BoC maintains the overnight interest rate within the upper and lower limit of the operating band of its Corridor System. The aggregate demand for settlement balances in the Corridor System is a downward-sloping inverted *S*-shaped non-inelastic but close-to-perfectly elastic curve. It is assumed that the supply of overnight funds, via the SLF, is perfectly inelastic, since the BoC is the de facto central bank of Canada. Demand for settlement balances is most elastic at the intersection between the demand and supply of settlement balances.

Figure A

As the overnight interest rate falls (i.e. closer to the deposit rate), the demand for settlement balances rises (theoretically to infinity). The BoC can then target an overnight interest by controlling the bank rate and deposit rate. Thus the BoC simply has to supply enough excess settlement balances to the Corridor System in order to drive the overnight rate down to the lower-end of the corridor.

Following the 2007-8 financial crisis, spillover effects from lower equity valuations globally meant that the BoC needed to made three adjustments to its Corridor System. The BoC made adjustments without the need to implement non-conventional monetary policy, unlike most other advanced economy central banks.

- The BoC set the target overnight interest rate at the bottom of the operating band, as opposed to the midpoint of the operating band.
- The operating band of the Corridor System was narrowed from 50 basis points to 25 basis points.
- The BoC expanded the target daily level of settlement balances in the Canadian Large Value Transfer System (
**LVTS**) from $25 million to $3 billion.

The first two adjustments meant that the Corridor System set an effective lower bound (**ELB**) on the overnight interest rate while the third adjustment created competition among LVTS participants in loaning out surplus funds at the end of each business day to reduce the incidence of cash hoarding. These three adjustments collectively pushed the overnight interest rate down and closer to the lower boundary of the Corridor System.

Figure B (below) is a modified illustration of the Corridor System by Nellie Zhang, which draws from Whitesell’s static model of Corridor Systems. Whitesell’s model assumes that a financial institution (involved in the demand for settlement balances) has an approximate idea about its target end-of-day funds position, *T*, but not with perfect certainty. Whitesell assumes two aspects of *T*: Firstly, a random liquidity shock, *E*, is included, which accounts for events that occur through out the day and out of the target settlement; and the stochastic estimation error is assumed to have a zero mean, such that E(*E*) = 0.

Figure B

The financial institution must, with no prior knowledge of the random liquidity shock, look to minimize the following two costs:

- The opportunity cost of depositing excess balances at the BoC, since you could earn a greater return at the market rate, modeled as (
*T*+*E*) x (*i*– (*i**+*s*), where (*T*+*E*) > 0. - The additional cost of borrowing reserves from the BoC to fulfil a negative settlement position, as opposed to borrowing from the market, modeled as (
*T*+*E*) x (*i**+*s – i*), where (*T*+*E*) < 0.

While noting the following:

*s*is equal to half the operating band, i.e. (bank rate – deposit rate) / 2, assuming the upper and lower limit of the operating band is symmetrically distributed.- Model assumes:
- A financial institution is indifferent at the margin between dealing with the BoC and trading with other financial institutions in the market; and
- The lending and deposit facilities offered by the BoC are perfect substitutes for the corresponding inter-bank transactions in the overnight market. The actual market rate may carry a premium reflecting collateral cost and/or credit risk adjustments.

Zhang’s illustration thus keeps the central bank’s target overnight interest rate, i*, set between the Bank Rate and Deposit Rate, and illustrates the dynamics of the demand and supply for overnight funds, a central assumption to Whitesell’s model.

Whitesell’s model makes the following three assumptions:

- The optimal amount of settlement balance for a finandcial institution is the inverse of the optimal amount of settlement balance for a typical financial institution, modeled as:
*F*(-*T**) = [1/2 + (*i*–*i**)/2s], where*T**is the optimal settlement balance and*F*(.) is the cumulative distribution funtion of the random liquidity shock (*E*). - The aggregate demand for settlement balances in the system is obtained by summing the individual demand over all participants with access to the central bank liquidity facility,
- Modeled as: , where central bank liquidity facilities are indexed by
*j*and aggregate demand in the Corridor System is*D*.

- Modeled as: , where central bank liquidity facilities are indexed by
- The random liquidity shock (
*E*) is assumed to be normally distributed, with a zero mean and non-zero variance,- Modeled as: , where is the uncertainty of end-of-day liquidity shock of the financial institution
*j*and Φ is the cumulative distribution function for the normal distribution.

- Modeled as: , where is the uncertainty of end-of-day liquidity shock of the financial institution

As illustrated in Figure B, the S-shaped demand curve flattens near both boundaries of the corridor since the demand for settlement balances can theoretically be positive or negative when the overnight interest rate asymptotically approached the Bank Rate and Deposit Rate. This occurs while the supply of overnight funds remains perfectly inelastic. Thus, as the demand curve flattens near the boundaries, the deposit rate widens and the interest-rate elasticity of demand decreases near the target interest rate.

Widening of the deposit rate is illustrated in Figure B as the vertical spread between *D *to *D¹ *or *D* to *D**²*, such that the Deposit/Bank Rate spread widens from *s* to *s*¹, such that *s*¹ > *s*. However, this assumes that random liquidity shocks have a symmetrical, which is difficult to hold in real application. As a result, the BoC then must provide a ‘small’ positive amount of settlement balances in the system, rather than near zero.

This is one advantage of adopting the Corridor System: demand for settlement balances should theoretically be zero so long as the target overnight interest rate is at the midpoint of the operating band. Conversely, if the central bank does not wish for zero settlement balances, it can, as a reaction to the 2007/8 financial crisis target the overnight interest rate at the bottom of the operating band and, by expanding settlement balances in the LVTS, create the incentive for financial institutions to flood the economy with excess funds and provide monetary stimulus.

That said, one drawback associated with a Corridor System is the realization of near zero settlement balances. This is difficult to achieve due to the intricacy of guaranteeing symmetrical opportunity costs around the target interest rate, given the imperfect substitutability between settlement balances and the private supply of broad liquidity, transaction costs and frictions in the overnight market.

Settlement balances are not completely equal alternatives to the private supply of broad liquidity since interbank loans are, in general, uncollateralized, whereas settlement balances simply require high liquid assets. This results in a material spread between the actual market rate and Bank Rate due to the premium for collateral costs and/or credit risk adjustments. This occurs due to:

- Transaction frictions and costs are associated with re-estimating the demand curve; and
- A limited number of LVTS participants means that the majority of financial institutions who do not have access to the LVTS end up borrowing overnight funds from LVTS participants at a higher interest rate than the prevailing interbank lending rate.

Empirical analysis of the incidence of a negative spread between average overnight interest rates and the target rate **suggests that the exclusivity of LVTS participants is a factor in the incidence of negative spreads**. The limited eligibility of participating in the central bank deposit facility leads to a segmentation in the overnight funds market, thereby giving financial institutions, who have eligibility to extract borrowing rates lower than policy deposit rates, generate a spread from lenders without dual choice. Since the demand curve is increasingly interest-inelastic near the target rate, a small error in the estimation of settlement balances can have a material impact on the overnight interest rate.

The inability to guarantee symmetrical opportunity costs around the target interest rate is a potential explanation to why the Federal Reserve has not adopted a Corridor System. In light of the fact that central banks of Australia, Canada and New Zealand adopted the Corridor System in the late 1990s, with the Bank of England adopting mid-2001, the United States is perhaps the last industrialized country yet to adopt such a system.

Empirical evidence compiled by the Bank of Canada and England respectively indicate that deviation of effective interest rate from target rate has been material, potentially a result of the increasingly complex and heterogeneous nature of financial markets. While the process of implementing a new monetary policy framework in the US would not only be long and arduous, legislative proposals and forming of new laws could stall the shift in framework. Therefore, the primary consideration for a switch in monetary policy framework of the Federal Reserve would need to consider whether a new framework could minimize the deviation between the target Funds Rate and effective Funds Rate.