Negative Interest Rates
By Callum Renton, Roberto Patiño, Sally Yang and Annie Wang
Introduction
We have chosen to investigate the contemporary phenomenon of negative interest rate policies (NIRP). Intuitively, the notion that a creditor is willing to pay a borrower for the pleasure of lending to them seems nonsensical. In practice, this has been proven to be an effective policy without wrecking havoc in financial markets. It is a policy undertaken extensively at the ECB, tempered at the Bank of Japan, and on the cards at both the BoE and Fed. We seek to understand how negative interest rates operate in practice, their implications for bond pricing, and criticisms of the policy.
Beyond the Central Bank
How have negative interest rates expanded into economies?
Previous articles have explained the transmission mechanism. Under the floor system, when the central bank temporarily cuts its deposit rates, defined as the money the central bank pays to commercial banks for depositing with the central bank, the market interest rate falls too. This induces commercial banks to lend more to consumers and investment projects instead of simply depositing at the central bank. Demand for such loans also increases since the interest rate is essentially the cost of borrowing. As consumer spending and investment in the economy increases, price levels get pushed up, resulting in inflation.
The conventional mechanism is generally expected to function in a similar way even if interest rates become negative. However, there is concern that commercial banks are not “passing” the negative bank rates to retail depositors. They may be reluctant to start charging for deposits because people can just withdraw their money and hold cash as an alternative. Essentially, the zero lower bound for deposits is thought to persist. This may harm the profitability and hence lending activity of banks with a particularly high proportion of deposits. The details and validity of this argument are further explored in the last section of this article.
One major benefit of lowering interest rates into the negative region is the alleviation of liquidity traps. After 2008, even though the ECB set their policy rates to or near 0%, investments, consumption, and inflation remained anemic for several years afterwards. This is due to economic pessimism, expectations of deflation, a preference for saving, and, very importantly, a preference for cash over holding bonds: Previously, 0% was considered to be the lower bound for interest rates. Thus, when interest rates are zero, investors will expect interest rates to rise and the price of bonds to fall some time in the future. This guarantees them a loss if they buy bonds now. As a result of their level k thinking, many investors would rather keep cash savings than hold bonds when interest rates are zero. By lowering interest rates into the negative territory, the central bank ensures that the zero lower bound no longer constrains market expectations and restores the effectiveness of monetary policy.
However, an effective lower bound likely still constrains market expectations. This effective lower bound is tied to the cost of storing cash. Interest rates cannot be cut forever; below a certain threshold, more people will be inclined to withdraw their deposits and store their money as cash. By adopting negative interest rates, the central bank may just have postponed the liquidity trap. Near the effective lower bound, people may again start expecting interest rates to rise in the future, causing the trap to be reintroduced. In the long run, having negative interest rates near the effective lower bound may “bake in” low long-run inflation and zero inflationary expectations. Trillions of government bonds are now trading at negative rates globally. However, none of the central banks that attempted negative rates seem to have measurably improved economic performance.
The abolition of cash has been argued as a way to improve the effectiveness of the transmission mechanism with negative interest rates. The central bank can attempt to create and encourage the use of its own cryptocurrency. Instead of holding cash, people will hold debit cards with the central bank and make payments with it. One example is the E Krona. If cash, which is the zero-interest-rate alternative to depositing with a commercial bank, is abolished, there will no longer be an effective lower bound. The central bank may then be able to more effectively influence consumer borrowing and spending through adjustments of negative interest rates.
However, this assumes that more extreme negative interest rates are a sensible and necessary policy solution to economic downturn. Negative interest rates punish savers and threaten future income, implying that it may be more sensible to “recharge” monetary policy through fiscal expansion.
Bond Pricing with Negative Interest
Negative interest rates have noteworthy implications on how standard bonds are priced and executed.
Introduction to Bond Pricing
Typically, bonds are priced as the present discounted value of future returns. For example, if we expect to hold a bond for one period, we would anticipate the return to be the resale value of the bond at the next period, plus any payment owed in interest for holding the bond.
For a bond that pays out Xt at time t, with a yield-to-maturity of i_t, maturing at time t+m, this reduces to:
Lowering rates to negative will lower yield and increase bond prices. This negative correlation is also implied by the asset pricing equation.
Bond Pricing with Negative Interest
In general, we know that lowering interest (and hence effective interest) will increase the price of bonds by extension of the previous logic. However, the effect of negative interest rates on specific bond pricing depends on the type of bond:
- For zero coupon bonds, such as most government bonds, the price will rise and the negativity will result from the final repayment being less than the initial loan.
- For non-zero coupon paying bonds in the secondary market, a negative yield would also imply that the price of the bond will rise, but any negative coupons must be outweighed by the present discounted value of the final repayment.
- Alternatively, for a non-zero coupon paying bond in the secondary market, a negative yield could be reflected by the initial loan exceeding the value of total repayments. The price of the bond would still rise, but we do not require that a final repayment outweighs any intermediate payments.
- For perpetuity bonds, the formula would suggest a negative bond price. This seems unreasonable and is unlikely to occur in practise for intuitive reasons explained later.
A negative interest rate potentially implies a negative coupon depending on the conditions of the bond. This would imply that bond holders pay the issuer each period in exchange for the bond.
If we define the absolute value of this payment as:
Then we can only rationalise this by ensuring a final maturity date where the face value of the bond (M) is paid back to the lender:
We know that bond prices are not negative, therefore it is the case that the component of the final payment outweighs the absolute value of the negative component, resulting in a positive bond price.
The perpetuity bond case cannot be computed as it has no maturity by construction. In this case, the asset pricing equation breaks as the price would be explosively negative. More intuitively, no agent would lend to another, paying them for the privilege each year ad infinitum.
Investors are still willing to hold bonds with negative yields due to the following reasons:
- Central banks still purchase bonds, as it will decrease borrowing costs and stimulate the economy.
- Institutions and investors which buy bonds must follow rules dictated by the banks, pensions funds or insurance companies who provide the bulk of the capital. Regulators force some clients to buy certain assets: banks can only buy liquid assets. Therefore, the buy side could effectively force some investors to buy negative yielding debt.
- If investors choose to hold cash, it may have a negative interest rate by depositing it at banks due to a negative base rate. A highly liquid but slightly negative yielding government bond would therefore look more attractive by comparison.
- Government bonds are safe compared to riskier assets. As fear grows, demand for haven assets increases. So buying safe bonds is a risk-averse strategy. Although there is loss of money for that safety, the loss can be seen as preferable if other asset classes tumble. Therefore, the issue of negative yields on government bonds may be overshadowed by their proven ability to rally during times of market distress.
Negative Yield Bonds in Practice
Negative interest rates are a reality in Japan, in every Eurozone country, and used as a policy instrument by the European Central Bank (ECB). We investigate whether the relationship between bond yield and price hold in reality:
Visually, the relationship holds for all bonds, with a (naive) negative relationship. Interestingly, this relationship does break at particularly low yields where price seems to fall as the yield decreases. However, this is likely to be complicated by short-term factors as the most negative yield bonds are short-term.
Sustainability of negative interest rates
We provide three assessments for the long-term efficacy (and feasibility) of maintaining negative interest rates in the long-term.
1. Banks might not transmit the negative rates to customers, shifting to more risky assets while reducing overall lending.
In a 2019 paper, Heider et al. suggest that negative policy rates may reduce, instead of increase, lending by banks. This may occur because most banks are reluctant to pass on the negative bank rates to depositors. The logic goes that if banks start charging people for depositing with it, people may simply withdraw their deposits and hold cash, which has zero interest rates, as an alternative.
In this sense, negative interest rates may particularly harm the profitability of banks with a very high deposit-to-asset ratio: not only are they being charged for depositing money with the central bank, they also still have to pay people for depositing money with it. As a result, the affected banks may be less able to supply credit to investment projects and consumers who want to borrow, resulting in an overall reduction in investments in the economy. Heider et al. finds that after the European Central Bank adopted negative policy rates in mid-2014, euro-area banks with a higher reliance on deposit funding have been taking more risks and lending less. This would suggest that negative interest rates could post a risk to financial stability, if lending is mostly done by high-deposit banks.
In the two previous tables (from Heider, 2019), we can observe the rise of the volatility of firms financed by banks following NIRP, reinforcing the argument of increased financial risk, and the reduction in syndicated loans in the Eurozone after NIRP, supporting the claim that negative interest rates might reduce lending rather than increase it.
However, most banks do appear to be passing on the negative bank rates on deposits held by corporations and institutions. Since late 2019, some banks in Switzerland also began charging negative interest rates on wealthy individual clients with high savings. Intuitively, this may be related to the suitability of holding cash as an alternative: while an ordinary Joe like you and me may be able to chuck our lifetime’s savings into a single strongbox, this is much harder and less flexible for rich people and corporations that buy and sell millions or billions of pounds at a time. While the high-net-worth depositors in Switzerland have an incentive to withdraw their deposits and store their money as cash or gold, the number who actually did so so far has been “limited” due to difficulties in securing a storage and insurance plan. Depositing money with a bank also has other benefits, such as credit that can be used to borrow against.
There is also evidence that negative rates affect a growing proportion of the deposits held by non-financial corporations, suggesting that the pass-through associated with negative policy rates has increased gradually over time. This induces firms to decrease their cash holdings through investments. The standard monetary policy transmission mechanism may still hold up.
The validity of the transmission mechanism under negative interest rates remains an open question. After all, negative interest rates are a very recent policy development. While there is growing empirical literature on its effects, a general consensus has yet to emerge.
2. Banks might be negatively affected if they rely on interest payments (eg. German Banks) & the effect on foreign exchange rates and yield curves.
In a paper by Eisenschmidt and Smets (2019) titled Negative Interest Rates: lessons from the Euro Area, the authors analyse the case of Germany, where there is an importance reliance of banks on interest income, making them more prone to suffer negative profitability from negative interest rates, which could in turn reduce lending as argued by Heider (2019). They also claim German banks are a representative example of the euro area:
“To further explore the pass-through of negative rates to bank deposit rates, we zoom in on the case of Germany, the country with the lowest sovereign yields and the highest level of excess liquidity of all euro area countries. While in many other euro area countries bank deposit rates had significant room to decline in response to the Negative Interest Rate Policy (NIRP) mainly due to higher bank deposit rates when negative rates were introduced, deposit rates in Germany were already close to zero at the beginning of the NIRP (…). Furthermore, Dombret and others (2017) argue that low interest rates pose a particular challenge to German banks, as they are especially reliant on interest income. The German case may therefore be considered most representative for studying what a steady-state pass-through of negative policy rates to bank deposit rates looks like.” Eisenschmidt and Smets (2019)
Critics have not only pointed out potential risks to the financial sector with reduced profitability and lending, but also increased exchange rate volatility as a result of the introduction of negative interest rates. In the same paper by Eisenschmidt and Smets, the authors quote the work of Gräb and Mehl (2015):
“Gräb and Mehl (2015) find that exchange rates of countries with negative policy rates tend to react more strongly to changes in their corresponding bond yield differentials vis-à-vis the U.S. For the euro area, their estimates suggest that a cut in the deposit facility rate by 20 bp is associated with a depreciation of the euro against the U.S. dollar which is around 0.5 percentage points larger in negative territory than in “normal” times. Overall, their empirical results suggest that negative interest rates make exchange rates more elastic to shocks.” Eisenschmidt and Smets (2019)
3. Are negative inflation rates always expansionary? There is mixed empirical evidence.
In traditional non-negative monetary policy, a decrease in the nominal interest rate of a central bank is used as an expansionary tool, to increase lending and foster economic activity, which in turn should increase inflation. However, there is mixed evidence on the assumption that this phenomenon holds for negative or extremely low territory. This could be due to the reluctance from commercial banks to pass on the negative interest rates, forcing a drop in profitability and a drop in bank lending. This friction could turn the reduction of nominal interest rates, counterintuitively, into a contractionary monetary policy.
This idea was discussed by Brunnermeier and Koby in a 2018 article titled The Reversal Interest Rate. They argue that reducing nominal interest rates could have contractionary effects because of the frictions in the banking sectors and the reductions in profitability as described before. However, they make it clear this “reversal” interest rate is not necessarily zero, which could ease some concerns over negative interest rates:
“Importantly, the reversal interest rate [the rate at which cutting interest rates reduces rather than increases lending] is not necessarily zero. Hence, unlike what some commentators suggest, negative interest rates are not fundamentally different” (Brunnermeier and Koby, 2018)
The article from Heider (2019), mentioned before, also supports the idea of the contractionary effect of very low interest rates, referring to empirical evidence on syndicated loans in the EU.
On the other hand, there exists empirical evidence on the expansionary nature of negative interest rates in the Eurozone. In a paper from Demiral et al. from 2019 titled “Negative interest rates, excess liquidity and retail deposits: banks’ reaction to unconventional monetary policy in the euro area”, the authors argue that NIRP has actually been expansionary in the Euro data:
“Using confidential bank-level data covering around 70% of main assets and 80% of total loans of euro area banks in a sample running until March 2018, we find that NIRP has been expansionary by inducing highly-exposed banks to increase their lending activity in an effort to mitigate the adverse impact of NIRP on their profitability.” (Demiralp et al., 2019)
The authors attribute the discrepancy in the results to the notable difference in sample sizes with Heider’s paper in particular, by pointing out the small share that syndicated loans represent of commercial loans in the EU:
“Our different result partly reflects the much wider coverage of our sample: Syndicated loans account for only 3% of euro area loans, whereas our sample includes the vast majority of euro area bank loans, including syndicated loans.” (Demiralp et al., 2019)
This article is written by: Callum Renton (Head of Europe Monetary Policy Research), Roberto Patiño (Research Associate, Europe Monetary Policy Research), Sally Yang (Research Associate, Europe Monetary Policy Research) and Annie Wang (Research Associate, Europe Monetary Policy Research). This article is reviewed by Jason Jia (VP, Monetary Policy Research).