Current location - Loan Platform Complete Network - Bank loan - Why can the "upside down" of deposit certificates and MLF interest rates continue?
Why can the "upside down" of deposit certificates and MLF interest rates continue?
Authors: Chen, fixed income analyst of Pacific Securities.

Since the RRR cut interest rates in July, 20021,the yield of certificates of deposit has continued to drop sharply. At present, the primary issuance of 1 has dropped to 2.73%, which is significantly lower than the MLF interest rate of 2.95%.

Figure 1 1 certificates of deposit continue to be issued, which is significantly lower than the MLF interest rate.

There have been two common sayings about 1 year certificates of deposit in the market:

One of the market viewpoints: 1 year certificates of deposit should fluctuate around 1 year MLF interest rate, that is, 1 year MLF interest rate is the center of 1 year certificate of deposit interest rate;

The second market view: when the interest rate of 1 year certificate of deposit is lower than that of 1 year MLF, it means that the market interest rate is low and the bond market should be treated with caution.

According to the above two mainstream market views, we can only get out of this year's bond bull market (note: we disagree with the above views and put forward "the debt cow has arrived" in March):

Since April 28th, the yield of 1 year certificates of deposit has been lower than MLF interest rate 1 2.95%, while the yield of 10-year government bonds has dropped by 28bp since April 28th.

Since June 30th, the yield of 1 year certificate of deposit is lower than 1 year MLF interest rate 10bp. Since June 30th, the yield of 10-year government bonds has also decreased 16bp.

Facts show that the above two mainstream views on certificates of deposit are problematic.

So what's the problem? What is the relationship between 1 certificate of deposit and 1 MLF?

First of all, it is necessary to clarify why the market generally thinks that the MLF interest rate of 1 year should be the center of the deposit certificate interest rate of 1 year.

The simplest logic is that both 1 year certificates of deposit and 1 year MLF are ways for banks to supplement 1 year liabilities and should be replaced. There are two logics for the higher-level explanation:

First, according to the central bank's monetary policy implementation report in the second quarter of 2020, it is clearly pointed out in the paper: "As a medium-term policy interest rate, the medium-term lending convenience rate is the center of the medium-term market interest rate operation, and the market interest rates such as government bonds and interbank deposit certificates fluctuate around the medium-term lending convenience rate."

Second, the OMO interest rate, that is, the policy interest rate, is really the center of DR007.

It is very obvious in 20 19. Although the OMO interest rate at that time was not an explicit policy interest rate, it had already played the role of an implicit policy interest rate, and DR007 fluctuated around the OMO interest rate.

Figure 2 The OMO interest rate of 2019 is already the center of DR007.

First of all, it must be clear that the above two points are indeed correct.

Based on the above two points, the market can easily draw the following inferences:

Since the central bank has made it clear that the OMO interest rate, which is also the policy interest rate, is really the center of DR007, then the 1 year certificate of deposit should not deviate from the MLF interest rate for a long time.

But this is not the case. First, let's review the 20 19 bond market.

In the whole year of 20 19, the MLF interest rate was about 3.3% (once dropped to 3.25%), and the fluctuation range of 1 CD yield was 2.95-3.3%, which happened to be the upper limit of 1 CD interest rate, but not the center.

Figure 3 20 19 MLF interest rate 1 is the upper limit of 1 certificate of deposit yield, not the center.

The MLF interest rate of more than one year 1 year is the upper limit of the deposit certificate interest rate 1 year, and the deposit certificate of 1 year can significantly deviate from the MLF interest rate. Why on earth is this?

We have studied this issue for a long time, and we have also read various statements of the central bank and various explanations of the market. Finally, we think the conclusion is:

First, short-term interest rates are more influenced by policy interest rates, while long-term interest rates are more determined by the market, which conforms to the basic principle of interest rate marketization (the central bank determines short-term interest rates, and the market adds term spreads, credit spreads and liquidity spreads to short-term interest rates to form other yields).

Second, because the OMO interest rate is a short-term policy interest rate, which is similar to the target interest rate of the Federal Reserve, the corresponding market interest rate (DR007) must closely revolve around the OMO interest rate, that is, the center of DR007 should be near the OMO interest rate.

Thirdly, the MLF interest rate is a medium-term policy interest rate, and the corresponding term is longer, so the market plays a greater role in it: the policy interest rate is more of an anchor, and the corresponding market interest rate (1Y deposit receipt yield) can fluctuate freely within a reasonable range, and there is no need to have a half-year or one-year average with the MLF interest rate.

Fourthly, from the operating frequency, it also supports the above conclusion, that is, OMO operates every day because of the need to accurately control DR007, while MLF operates every month, which has much less intervention on the yield of 1 CD.

Fifth, another policy significance of MLF interest rate is that MLF affects LPR, which in turn affects the loan market. MLF's policy benchmark for loan interest rate is actually bigger.

Among them, the relationship between 1 certificate of deposit and 1 MLF is actually similar to the "dog-walking theory": "When walking a dog, the dog sometimes runs in front of people and sometimes runs behind them, but generally it will not be too far away from people."

The most important point here is that dogs rarely stay around people, but run around, that is, dogs can deviate from people's position for a long time within the length of the rope.

The same is true: 1 year certificate of deposit can be lower than 1 year MLF for a long time, such as 2019 for the whole year; 1 year certificate of deposit can also be longer than 1 year MLF, such as from September 2020 to April 20021year.

Figure 4 The deposit interest rate can be lower than MLF or higher than MLF for a long time.

What is the reasonable range for 1 year certificates of deposit to deviate from 1 year MLF interest rate? We generally think it is around 35bp.

There are three main foundations:

First, for the whole year of 20 19, the minimum interest rate of 1 certificate of deposit was 2.95%, which was just 35BP lower than the MLF interest rate 1 at that time.

Second, in the second half of 2020, the highest interest rate of 1 certificate of deposit was 3.3%, which was just higher than the MLF interest rate of 1 at that time, and the range was also 35bp.

More importantly, the third argument is as follows:

June 30, 2020 165438+ added by the central bank 1 annual MLF (usually MLF is on June 5438+05). The central bank clearly stated in the monetary policy implementation report: "MLF interest rate, as a medium-term policy interest rate, plays a guiding role in the medium-and long-term interest rates in the market. In June, the interbank deposit interest rate and the national debt yield rate were 5438+02.

In other words, MLF was added to165438+1October 30th, just to increase the operating frequency and guide the market interest rate at that time to be close to MLF.

Let's look at the deposit interest rate at that time. 1 1/In late October, the deposit interest rate of 1 was just around 3.3%, which was just 35bp higher than the MLF interest rate of 1 of 2.95%.

In other words, the deviation of 35bp has reached the acceptable boundary of the central bank.

To sum up, we think that the relationship between 1 certificate of deposit and 1 MLF is as follows:

First, the yield of 1 year certificates of deposit should fluctuate within the range of "1 year MLF interest rate of 35BP";

Second, the yield of 1 certificate of deposit can be lower than MLF interest rate for a long time, or higher than MLF interest rate for a long time;

Thirdly, there is no mean regression of "deposit certificate interest rate -MLF interest rate" spread on the monthly time scale; This is why it is not feasible to look at the degree to which the deposit certificate interest rate deviates from the MLF interest rate.

For the current certificates of deposit and bond markets, the conclusion is simple:

The MLF interest rate of 1 year is currently 2.95%, and the range of 35bp is 2.6-3.3%, that is, the yield of 1 year certificates of deposit can be reduced to 2.6% at the lowest.

At present, 1 certificate of deposit is 2.73%, which is still in a reasonable range, and it may not converge to MLF interest rate in the short term, and there is still a small downside (although the space is really not big).

Based on 1, the deposit interest rate is significantly lower than the MLF interest rate, which is not in line with the intention of the central bank and the historical experience of 20 19 to bearish on the bond market (the deposit interest rate is significantly lower than the MLF interest rate for a long time, but not more than 35bp).

We can analyze the yield of national debt in the same way. Except for 20 17, when funds were very tight, and March-May, 2020, the yield of 10 was close to that of 1 in most historical periods.

At present, the yield of ten-year treasury bonds is around 2.92%, which can be lowered to 2.6% according to the certificate of deposit benchmark. Even if 10-20bp is added, the yield of 10-year treasury bonds can be lowered to 2.7-2.8%, and there is still room for decline.

Another way is to price the bond market according to our "revaluation": the previous market priced the bond market at 3.0-3.3%, but this pricing itself is wrong, suggesting that the central bank will raise interest rates and return to the assumption before the epidemic; The central bank can't see the possibility of raising interest rates in the foreseeable future, so the bond market yield should be priced at 30bp lower than 20 19 (the interest rate cut last year), that is, 2.7-3.0%, and above 2.9% is still at a high level.

If there is a rate cut in the second half of the year (export, industrial growth, PPI trend downward), then the range will be further reduced.

To sum up, the conclusion of this paper is:

Both 1 and 1 certificates of deposit yield are reasonable within the range of "MLF interest rate of 35bp". At present, 2.73% is far from the lower limit of 2.6%.

2. Because there is no regression of deposit interest rate to MLF interest rate on the monthly time scale, it is not feasible to see the extent to which deposit interest rate deviates from MLF interest rate.

3. According to our "revaluation" of the bond market, the range of 3.0-3.3% previously expected by the market should be adjusted to 2.7-3.0%; For the whole year, we think that the yield of 10 bond is expected to drop to 2.7%, which is based on our "gravity" (the average return of the spread between the bond market and the money market).

4. The short-term bond market is expected to be extremely chaotic, and the short-term trend has completely depended on market sentiment. There is the possibility of unilateral decline in July-August similar to 20 16 (asset shortage, self-realization of market expectations), and there are also short-term predictions similar to September 20 19 and June 20021(the market is ahead of the central bank and excessive), which is basically equal to gambling, and coping is more important than predicting market sentiment.

5. In combination with the long-term and short-term, we think it is still necessary to continue to do more, and we can make appropriate profits by doing more heavy positions in the early stage (according to the "debt cow has arrived" we put forward in March, we should put heavy positions at that time, and we should put heavy positions in mid-June at the latest), and light positions still need to be replenished; If the bond market can be adjusted, it is an opportunity to increase positions.