A Long Term Interest Rate Outlook

To come to a long term interest rate outlook we should first look at interest rate history and the outlook for growth.

30 years ago our western/developed economies were experiencing higher inflation and growth than today. The inflation was tamed by a combination of higher interest rates and some slowing in government deficit spending. As inflation reduced, nominal interest rates fell, but real (inflation adjusted) interest rates did not reduce anywhere near as quickly.

Over the past 30 years long term interest rates have fallen from ~12% pa to ~1 pa. So asset values have effectively had a boost in the price to earnings multiple from 8.5 times to 100 times if we simply use the inverse of the long term bond rate, However equities, being volatile and prone to occasional crashes, command a premium over bonds to attract purchasers, so let's make the growth in PE multiples from 10 times to 33 times.

This fall in interest rates has also reflected slowing of real growth rates on a per capita basis in most developed countries. Nominal GDP has grown through the effects of immigration, inflation and productivity.

These falls in interest rates did not of themselves cause major realignments of currencies, because the governments of other western countries were doing the same things. While the currencies were generally losing purchasing power (generally measured in relation to a basket of consumption goods) at similar rates the relative values of the currencies tended to relatively remain stable, except for other reasons eg Australia and it's post GFC resources development boom.

These interest rates were driven by giving central banks an appearance of independence and a mandate which typically includes goals of maintaining full employment and keeping the value of the currency reasonably stable over time at a target rate of inflation (loss of consumer purchasing power).

In fact, central banks are not wholly independent of governments as the central bank must respond to the impact on the economy of the change in spending by governments. If governments spend too much inflation breaks out and central banks raise interest rates and if governments don't spend enough the private sector loses spending power and the central bank cuts interest rates to increase spending power of most businesses and consumers (but reduces the spending power of those dependent on interest income, which is typically older retired people).

The relatively consistent increase in asset values brought about by falling interest rates over a long period creates a golden era for governments, democratic and autocratic, because everyone is happy to see the value of their assets going up, whether it be their home, retirement savings, share market investments or investment properties. The increase in value of housing also provides a way to assist funding of retirements as retired people sell large homes in prime locations for employment and buy smaller homes, sometimes in less expensive areas where there are reduced opportunities for employment. The reduced employment opportunities do not affect retired "empty nesters".

So, given that background, what is the outlook for long term interest rates over the next 30 years?

Well, to achieve the same outcome of a golden glow of happiness in the population, why wouldn't governments continue with the same approach?

So expect to see one of three things:
1. interest rates keep falling through the 0 mark and moving increasingly negative over the long term 

2. increased deficit spending by governments to maintain full employment or
3. some combination of the above.

At present progressive governments tend to spend more to increase employment and conservative governments restrict spending to maintain a pool of unemployed to give employers more bargaining power and thereby force central banks to cut interest rates. This is often done under the guise of "stopping the blowout in government debt"

Japan has tended to increase deficit spending rather than reduce interest rates substantially below the 0 mark, especially at extreme times such as during a recession. Lowering interest rates does not send as powerful a signal that the economy and jobs will be protected as government showering money on the private sector. When governments are practicing what they call "fiscal responsibility" and the economy is slowing, cuts in interest rates are often sufficient to increase spending, but not when there is a severe economic or social disruption.

So over the next 30 years, I expect interest rates to continue to fall when conservatives are in power and will not be surprised to see the 0 bound breached more and more frequently until we enter a period of stained negative interest rates, even if immigration is able to be resumed Covid-19. When progressive governments are in power we will see expanded deficit spending and higher so called "government debt" (although much of it will be held by central banks and so be fake debt owed by one part of the government to another part of the government). Over the long period we will see the effect of alternating policies as we have seen over the 30 years. 

 

It is possible that in 10 years we could see long term bond rates at -5% pa, bank deposit rates at -6% pa, home loan rates at -4% pa, and for house prices and stock prices to have continued increasing at 5% pa and so producing capital gains for investors even from today's seemingly bubble prices.

John Hussman - two observations

I really like John Hussman's work.

But, and there's often a but, I have two major criticisms.

1. John largely ignores the power of sustained changes in long term bond rates and the trend which has driven large parts of the increase in earnings multiples incorporated into prices, the Price Earnings (PE) ratio.

2. John looks at stock market valuation largely in isolation from the need to save/invest surplus funds and so largely ignores comparison with other investment opportunities (other than occasional references to bonds or cash).

1."Permanent" changes in the required/assumed earnings rate drives changes in capital value

We have been in a 30 year bull market in bonds as interest rates have slowly fallen from the highs of the war on inflation to the present day occasional fear of deflation. It is important to recognise that what John has called "yield seeking behavior" is not new, it is just played out in an adjusted context.

Some simple maths. The inverse of the rate of income is effectively the price multiple applied to the income stream. There are lots of potential qualifiers to that statement but lets accept it as a broad generalisation. If a an assumed permanent rate of earnings is 10 then the implied capital value is dependent on multiple applied. So if one assumes 10% pa in perpetuity the value is the inverse of 10% which is 10. If the assumed required yield is 8% then the multiple becomes 12.5x. If it is 5% the multiple is 20x, 3.33% it is 33, if it is 2.5% then 40, 2% then 50, 1.5% then 66.67x, 1.25% then 80, if 1% then 100, if 0.5% then 200 . Note that as the required earnings rate approaches 0 the multiple gets higher and higher (it approaches infinity). If the required rate is only 0.10% then the earnings multiple is 1000.

So looking at the last 30 years, the 3 year average of long term US bond rates have fallen from about 12% to about 2.5%, which implies a change in the multiple from 8.33 (100/12) to 40. A perpetual bond earning $12 at a yield of 12% was worth $100 ($12 x 8.33) but now the same bond would be worth $12 x 40 = $480. The fall in long term rates from 12 to 2.5 has resulted in a 480% increase in the capital value. This is permanent if the change in interest rates is permanent.

What happens if over the next 30 years the 3 year average of long term bond rates falls by the same proportional amount as over the last 30 years. Over the last 30 years they fell from 12 to 2.5 or by 79.17%, so they are only 20.83% of the 12 they once were . So if they fall from 2.5 by the same proportion over 30 years they will be only 20.83% of what they are now, that is in 2050 long term rates would be 0.52% and the perpetual bond earning $12 would be worth $2,307. Now lets see how much of an increase that is in the bond price. It was worth $100 in 1990, 480 in 2020 and in 2050 it is worth $2,307. So how much did it increase over the second 30 years? You guessed it, by 480% again!

As an aside, we have recently all learnt a couple of things.
a) there is no 0.0% lower bound of interest rates, central governments can manipulate them into negative territory times, even so negative that consumer rates can also be negative:
https://www.theguardian.com/money/2019/aug/13/danish-bank-launches-worlds-first-negative-interest-rate-mortgage
b) no matter how much a government is ideologically focused on reducing taxation and welfare and reducing budget deficits and repaying the debt, when it really hits the fan they will run huge deficits to maintain stability, or at lest to dampen the turmoil. (See US, British and Australian governments fiscal response to Covid19
c) central banks and governments can do "whatever it takes" (Mario Draghi European Central Bank) when they so decide, more so if they are a sovereign issuer of their own fiat currency, and moreso again if that currency is a global reserve currency. Such countries don't have to tax to spend, but that is another topic.

 So let's look at our long term institutional settings. Inflation targeting has been all the rage among central banks, but complicated by also having to target full employment, and all the while having to do this without any control over fiscal policy. If fiscal policy settings are too tight then interest rates will have to compensate by getting lower, even turning negative. If fiscal policy becomes overly expansionist then inflation might take hold and cause interest rates to rise. So the future depends significantly on the dance between fiscal and monetary policy of governments.

John has done analysis of US demographics (beforeCovid 19 hit) and made estimates of real growth in productivity and concluded that there is little to drive real growth in income per capita beyond about 2% pa. That means that in the absence of some black or white swan there is  little to cause huge economy wide demand for money for capital investment other than to achieve lower labor costs or to replace economically obsolete assets. Lower migration means lower population growth means lower housing demand and lower growth in GDP, all other things being equal.

So the big question is why do stock market averages that include when we were on the gold standard, when we were on a fixed exchange rate standard, while we were fighting a war on inflation, while we were fighting either World War matter at all? Why is the past any indicator of where things should be? Why should the market revert towards where it was at those times?

There are some economic policies that could make US GDP change at a different rate such as allowing an increase in immigration and rebooting manufacturing by moving away from free trade and towards isolationism, but these policies would be contentious and impose other costs, direct and indirect on parts of society that they would otherwise not suffer, but the policies would benefit the stock market and owners of capital.

2. The value of the stock market will ebb and flow around the outlook for other mainstream investments.

John looks at the value of the stock market not so much in relation to other asset such as real estate, gold, cash or commodities but in relation to what it used to be at some earlier time, whether a boom of a bust. He has recently been projecting very low to small negative returns in the US stock market over the coming 10 or 12 years. But what returns will real estate have, or cash, or bonds, or foreign currencies or commodities such as oil or soybeans? Will the stock market provide a higher return than those assets or lower? There are many advantages to investment in major stocks including no property taxes, no vacancy factors, low transaction costs, ability to sell small parts of a holding over time to provide funds for living in retirement and they also form part of any assessment on where to invest. On the other hand there can be more volatility and greater short terms losses and often people tend to sell after they have endured much of the fall and then not get back in before the market gets above where they sold.

John's analysis is also restricted to US stocks and that is understandable but it also ignores about 46% of the global stock markets. Will Japanese, German, UK or Chinese stock markets be better investments than the US market? Will the USD depreciate against the EUR?

So John's analysis is not about macro asset allocation either geographically or between asset classes and so is only relevant for that part of your investment funds that you have already decided to allocate to US stockmarkets. But it does not help much with decisions on stock picking, sector weightings, which indexes to follow through ETF's. So it is of little help in making specific investments other than a decision in relation to US ETF's which largely track the S&P 500.

There are however, people watching all these different markets and crossing over between them, so no one very large market is going to have a much higher expected risk adjusted rate of return than any other such market for very long as investors and fund managers will begin selling in one market and buying in the one with the higher expected risk adjusted rate of return, It could take weeks for any premium to dissipate, or it could happen almost instantaneously, but it will happen. So although the US S&P might have projected returns of only 0.5% over the long term, that might be more than the expected return from holding cash. All asset prices that are valued based on a return are susceptible to adjustment as long term bond rates move to new seemingly permanent lower of higher rates, for the reasons discussed at  1. above

Some techniques of investing accentuate new trends because as they emerge from the activities of some investors others identify a change in trend and jump aboard and there are various systems for choosing when to jump into a trend,

So while the expected return from the sock market is low, investors are betting that they will get a greater return there than from other alternatives such as cash or bonds, or the UK market or Japanese market.

Thanks John for all your great work, which I have been reading for over 10 years.

Ethics of Triage

The great majority of triage decisions are based on the memory of a protocol in a time of great stress.  In real life it is never as simple as saying run over a dog to avoid hitting 4 people.

The base protocol is to save as many as you can by letting those most likely to die anyway die without any attempt to save them. In this way the lifesaving efforts are directed to those with the best chance of surviving longer term.

It was reported during March 2020 that in northern Italy as the health system was overwhelmed by Covid19 patients triage was for ventilators was based on age and that at one stage people over 60 were denied ventilators so they could be given to younger people who needed them and the allocation decision was being made by front line medical staff without benefit of a protocol as none had ever been needed before for this type of decision.

This can lead to long held feelings of guilt for the person making the decision and affect their willingness or ability to continue in their career.

To protect the mental health of the person making the decisions, a protocol is determined in advance and those making the final decision they are told that they are not making the decision to let someone die, they are merely following a policy determined by others that is fair and reasonable in difficult circumstances.

Some take all human consideration out of the decision by making selection of those who get lifesaving treatment by some form of lottery. Every patient in the hospital on a respirator and those waiting for a respirator are given a number. A means of "drawing lots" is used to see who wins or loses respirators. Respirators are given to those who win (or don't lose).

Another such way is simply first come first served. If you need one you go in the queue and if you are lucky someone with a respirator gets better or dies and you get there respirator. If you are unlucky you die before you get to the top of the queue and a respirator becomes available.

Both of these methods can result in saving the life of an old sick person with advanced dementia over the life or a newly qualified intensive care specialist, MBA, Ph. D. or budding Mother Theresa and this imposes a significant cost on society as a whole. There has to be a better way.

An economic rationalist would say what is the amount this person can contribute to production of valuable goods and services in the future versus how much they are expected to need support from the rest of society over their lives? Anyone in paid work would be saved in preference to someone not expected to do unsubsidised paid work  in future and even moreso if they are on any form of government benefit.

Age as a determinant can be an independently verifiable way to triage in a hospital setting and estimating age can be a "quick and dirty" way to triage, but with many errors at the margin. Those that look old are offered whatever palliative care is available if any and those that look much younger will get the ventilator and hopefully be saved (although a large proportion of patients who are put on an invasive ventilator (ie intubated) do not survive.

However some young people have diseases that limit their life expectancy severely, while some older people have long productive years of life in front of them.

Also, among older people there can be significant differences in life expectancy. Cancer and heart disease for example can mean a very low life expectancy for a younger person than an older healthier person has. Between two older people a smoker may have a lower life expectancy than a non-smoker  or a mature person with diabetes, hypertension, coronary artery disease and early stage dementia may have a lower life expectancy than an older person in good health.

So, if the goal is to maximise the overall benefit to society perhaps the expected disability adjusted life years is a better measure. But what about parapalegics? How is their disability taken into account? Does it matter if they caused it themselves eg by crashing a motorbike at high speed on the wrong side of the road? What about if skydiving? Or hit on a pedestrian crossing  in broad daylight by a car driven by a drunken drug addict that ran a red light in an unregistered vehicle?

Can people coming into hospital be given a point score on admission that takes all the circumstances listed above into account? How long would it take? how much would it cost to administer? Could you appeal? Could the dependants sue if the calculation was performed negligently and the person died for lack of a ventilator or an operation?

From the point of view of practical application the decision in accordance with the protocol has to be able to be made quickly and in accordance with the protocol by the front line worker. Society and the front line medical worker are both concerned that the perfect does not become the enemy of the good.

In a hospital setting, sometimes the decision can be made by the patient even if not conscious or having mental capacity through the application of a living will made previously by the patient. Examples are "Do Not Resuscitate" directions. There are also directions that the patient is not to be either put on or kept on life support for more than eg 5 days if they meet certain criteria. The criteria can be certain illnesses, prognoses or statuses. It could also be that certain other medical treatments are to be withheld based on certain criteria. A signed, witnessed living will and an enduring power of attorney copied to family members can make clear who is to make the decisions and in what circumstances they have discretion and in which circumstances the living will must be followed. This can relieve the front line medical staff of having to make any triage decision at all in most cases and also ease the making of difficult clinical decisions.

In many ways however the "best" triage decision is likely to be made by a senior medical practitioner who has thoughtfully considered all the above issues in advance and to relieve that burden age is a good but imperfect proxy for rapid estimation of societal benefit when overlaid over the clinical decision of who is most and least likely to survive in the medium term.

Fake Government Debt in Australia

It's fake debt!

There need be no debt crisis from Australian Commonwealth government rescue packages legislated to sustain the economy during the economic crisis caused by the Corona virus health crisis and response because it is fake debt.

There are 7 principles to understand.

1. When a commercial bank lends money it creates the money out of nothing (DR Loan to Joe Bloggs, CR Joe Bloggs cheque account, no outside or additional funding needed to make the loan. When Joe spends the money then, unless Joe spends it with a customer of the same bank, the lending bank may have to borrow from the general money market). (See the confirming Bank of England paper here)

2. A commercial bank can buy a bond, by creating the money from nothing (Dr Asset: Bonds, CR Commonwealth Government Treasury).

3. If it so desires the Reserve Bank of Australia ("RBA") and Commonwealth Government ('the Government") acting in concert RBA can make it attractive for banks to buy bonds from the Government and then sell them to the RBA, either by simple suasion or by making regulations that banks have to hold reserve assets such as Government bonds or credits at the RBA

3. When the Reserve Bank of Australia buys something like Commonwealth Government debt (Commonwealth Government Bonds) from a commercial bank, it does not inject money into the real economy, it simply recognises that the bank has increased reserves at the RBA (DR Asset: Government Bonds, CR Commercial Bank (eg Westpac) Bank Reserves

4. The RBA does not have to pay interest on reserves (but in some circumstances the government may make it pay such interest)

5. When the RBA makes a profit eg from interest on Commonwealth Government bonds, it pays that profit to the Federal Government Treasury as a dividend.

6 . Effectively the Treasury funds the interest it credits to the RBA as holder of the bonds from the additional dividend it receives from the RBA's additional profit. It is what is commonly called a "round robin". The money ends up back where it started.

7. When commercial banks get no interest on reserves at the central bank, their returns on assets fall a little compared to just before they sold them, but their profit will go up slightly as they will sell the bond at a slightly lower yield than that at which they bought it, effectively taking a small fee for their trouble.

So all this debt can effectively be at no net cost to the government and never have to be repaid.

This all works because the Commonwealth Government is a monetary sovereign that created and controls the Australian dollar as a fiat currency, controls the banks and controls the reserve bank and owns and runs the Australian Government Treasury

This does not work for state governments as they do not issue and control their own currency, instead they use the Australian Government's currency. Also the state banks do not control the RBA or the Australian Treasury.

It does work for other countries which are monetary sovereigns and have similar structures and institutions to the Australian Government eg USA, UK, Japan, China.

It does not work for countries that have surrendered their monetary sovereignty to some higher (in relation to money) authority. Primarily that is the countries that use the Euro as they ahve agreed to be bound by European institutions and to use a currency that the country does not control, the Euro.

The government could do this with more debt, it is simply a matter of political will, but there are some real limitations. If the Government gives people too much spending power it can cause asset or general inflation (and eventually hyper inflation). If the Australian government does huge amounts and other nations do none, then the Australian Dollar ("AUD") will fall in value relative to other countries' currencies.

Ending Corona Lockdowns

Spread of Corona Virus as of 7 April

While it is still early, we need to think ahead about ways to end lockdowns assuming we cannot totally eradicate the disease.

Remember there is no vaccine against AIDS and it has been with us for about 40 years. We cannot be sure that we will have a safe reliable vaccine for Covid 19 in 12 or 18 months. We cannot stay in lockdown for years. Lockdown will likely start to break down after 3 to 6 months for a variety of reasons, social, economic and psychological.

But there is a strategy for getting out of lockdown and self isolation with minimum deaths based on mortality rates by age group. 

Based on Australian data:
No deaths out of 1600 cases in the 20–29 age group
No deaths out of 830 cases in the 30–39 age group
No deaths out of 700 cases in the 40–49 age group
1 death out of 820 cases in the 50 to 59 age group

Australian Government modelling released 7 April shows very low hospitalisation and intensive care rates for people under 50, particularly for those under 30 (based on analysis of over 5000 cases)

I have not examined the younger age groups as the case numbers are too low to have much confidence in extrapolating the results from them at this time, but that is perhaps a great indicator of low transmission rates to young people or lack of health problems among them when infected such that few are being tested.

From those mortality numbers above we can see that the risk of death in these age groups is virtually zero provided the very sick of those infected have access to treatment including ventilators and provided that they have no other chronic illness when infected.

So we can gradually release all very healthy people in these age groups from lockdown and expect a death rate not much worse than seasonal flu provided the health system is not overwhelmed, but it will be a lottery as to who will die.

Over the course of the next say 2 months we would continue to build out hospital beds, ICU beds, PPE, ventilators and building our knowledge of what drugs help recovery and be that much closer to a possible vaccine and as current cases run off as a result of the lockdown medical staff and ffirst response staff could take breaks.

The sensible place to start release from lockdown is with all the very healthy 20-29’s who do not live with any over 50’s or people with chronic illness and release them all gradually over the course of 1 month. They should all be able to carry proof of age such as driver’s licence, official ID card or passport.

No one would have to leave isolation and healthy people eligible for release from isolation but living with very young, elderly or sick people would be encouraged to maintain the isolation of those people.

As release from isolation commenced bars, restaurants and shops staffed by this group could open with some protection for staff eg screens around service areas. Gradually members of this group would become infected. We know about 85–90% would need no treatment at all, maybe 10% would be hospitalised and 5% would need intensive care, but they would not all get sick at once so hospitals, intensive care, staff, PPE and ventilators would not be an issue.

As people recovered they would become immune and could then assist in relieving some medical staff and nursing home staff and take higher exposure jobs such as serving customers, particularly if we then have proper reliable antibody tests. This would also be the start of building herd immunity.

Depending on hospital, ICU and ventilator unused capacity, we would then begin to release the very healthy 30 to 39 year olds at a pace calculated not to overwhelm the medical system. More businesses would reopen or increase activity as more staff and customers became available. Herd immunity would continue to build. Economic cost of lockdowns would start to reduce, residential rents could start to be paid and slowly so would commercial rents.

Over a period of about 4 months most of the workforce will have been released from lockdown and herd immunity will be continuing to build. At no time would hospitals, ICUs or ventilators have been overwhelmed.

During the 4 months, we would expect to also have data as to releasing the 15–20’s. There is no reason to suspect that this group would not be able to be released subject to hospital/icu/ventilator/numbers.

In Australia we have had 5 deaths out of 900 cases for 60–69 age group.

At this stage (about 6 months) it may be that there could be release of over 60’s (again on a voluntary basis and with full explanation of the risks and with people with chronic illness such as hypertension, diabetes or coronary disease or other illnesses not leaving isolation). People would weigh up the risks for themselves after getting a personal medical briefing.

By 9 months about 60 to 80% of the population would have been infected. There would have been some deaths, some totally unpredictable and unlikely. Herd immunity would be at a stage where it reduced the chances of catching the disease and would be reducing loads on medical resources.

After this 9 month process, the over 70s and those with chronic disease would perhaps face difficult decisions, but they would have the benefit of reasonable herd immunity if they chose to leave isolation. They would also have the benefit of well developed treatment protocols and informal trials of many differing treatment regimes. It may be that infection isolated communities would be developed for such people so that they lived freely within a highly protected community.

There are some practical difficulties with this approach, but every approach has difficulties. Some people will cheat, but so long as they don't end the self isolation of someone else and there are not too many cheats it won't really matter. The main issue is for older people who spend more time in lockdown and feel they are denied economic opportunity. This could be eased by increasing the safety net payments as time goes on from the savings of less people being eligible or needing the safety net.

The issues of how long immunity lasts may still be with us in 12 months time but by then we may be within mere months of a vaccine. Time alone will tell.

(image from https://commons.wikimedia.org/wiki/File:COVID-19_Outbreak_World_Map_per_Capita.svg)