Is there a safe asset shortage?

Consider my previous hypothesis that so many yields have gone negative because of the insurance value of those relatively safe assets.  If that were true (true to some extent, I am not claiming that is the only factor behind the supply and demand curves), could a problem be solved by issuing more safe assets?  That could be done through more government spending or big tax cuts, either creating more government borrowing and thus more safe assets.

I often see it taken for granted that “selling more insurance” into the face of market demand would be a good thing.  And maybe so.  But it doesn’t follow in any simple way from theory, as is sometimes implied.  For one thing, this isn’t a straightforward market setting but rather there is a monopoly supplier that is not selling market outputs, and furthermore the private and social returns to either insurance or risk-taking are not in general equal.

So ask yourself which has a higher social rate of return?

1. How the government would spend marginal funds, plus relatively wealthy people buying more insurance

2. How private consumers would spend a tax cut, plus relatively wealthy people buying more insurance

3. Relatively wealthy people allocating more funds to riskier investments, and not getting that additional insurance

There is no way to tell, not from theory or for that matter from any kind of simple, straightforward empirics.  And from a social point of view, we therefore do not know if more safe assets should be put on the market.  There is of course a fourth possibility:

4. Because private investors enjoy more safe assets, they end up taking greater risks and the whole PPF shifts out

Again maybe, maybe not.  If it is true, a variety of subsidies to insurance markets, implicit or explicit, could improve welfare.  That is in contrast to Kenneth Arrow’s older argument that it is the risky investments that need to be subsidized and subsidized directly.  I say there is only so much “crowding in” to be had, especially during a time of near-full employment.

Another argument might be:

5. Selling more safe assets would raise nominal yields and make monetary policy potent again

I don’t want to go through all of that again, but suffice to say it had a better chance of being true in 2010 than today.  Here is John Cochrane on related issues.


The more you sell, they less safe they all become.

Let's suppose the US government initiated a $4 trillion stimulus for this year. The market buys it up with, basically, $4 trillion in cash when the dust settles. That cash gets spent and deposited in the banking system as someone's income for the year. So one group of people have $4 trillion in bonds as assets, another group of people have $4 trillion dollars in cash at the end of the year. What does the second group do with that cash to protect it in year 2? Do we initiate another $4 trillion stimulus to give them safe assets to invest in? Or just go back to the beginning- why was original $4 trillion dollars in cash less safe to begin with?

I think it is conventional wisdom that monetary stimulus creates more large money pools than fiscal stimulus would do.

I am wary of super large efforts but people I respect, like Bernanke, say that the fiscal lever has been under used.

What does "large money pools" have to do with it? Fiscal stimulus ends up as cash balances somewhere that would require another fiscal stimulus to provide more safe assets, if you believe the original argument about a shortage. The entire argument about a "shortage" just seems like bullshit in this light.

And if fiscal levers have been underutilized in the US, then Japan should be growing gangbusters the last 20 years. Indeed, we should be growing gangbusters- we have nearly doubled the federal debt the last 8 years, and have seen GDP grow a total of 1.2% the last year, and over the last 7, have seen the weakest recovery from recession since the Great Depression.

If the last 8 years have proven anything it is that money supply does not automatically flow to main street, and cause inflation.

Remember hyperinflation fears? Then the missing inflation conundrum?

Also, yes Japan has provided some lessons about fiscal policy. I think is is more conventional now to doubt infrastructure megaprojects, and there is renewed talk of helicopter drops.

Canceling college debt rankles me, but it is probably more a helicopter drop than rebuilding Tyler's favorite Zee bridge.

Why is the question always "Why don't we generate more safe assets?" instead of "Why are so many people forced to buy safe assets? This is crazy that even with negative nominal rates the large buyers are not disappearing." And while it's a good attempt, the insurance angle isn't significant enough to explain what's really going on. It fits with the daily market movements as expectation of low growth do drive rates down, but we know institutions are buying for other reasons.

The excessive demand for safe assets is largely driven by non-economic actors. This includes Basel III banks, QE central banks (some people see this variable as a supply factor), duration matching pension and insurance funds, SWFs and other foreign official purchasers. There are many sources of liquidity who cannot buy equities instead of bonds because they legally are not allowed to do so.

Right now it is safe assets that are being indirectly subsidized, and the signs are staring us right in the face as we see so many institutions forced to buy negative yielding bonds.

What percentage of Treasury issuance is held by US Basel III banks for capital requirements? Current treasury securities outstanding is around $13T, and around half of that is held by foreign nationals.

Even if you are correct that these requirements are driving the increased cost of insurance, so long as we have a financial system that would melt down if certain banks were to fail, it seems like forced insurance purchase by those institutions serves a purpose.

If I could borrow at negative rates I'd borrow as much as possible and sit on the cash I had to pay back, and spend the surplus on something useful. Why don't governments do that on a sufficient scale to make interest rates go back up to zero? Don't their finance departments know how to finance themselves? They could give tax cuts, or reduce their deficits by borrowing more.

As the sign says, Welcome to Ontario. More to Discover.

It is a wonderful way for democratic governments to consequence free.

The effect would certainly be self-limiting, but I think it is surprising how little effort smart policy leaders have put in naming best, first, uses for free money.

It is almost like people are human, and their values are sticky, so that debt is bad, even when it is free (or better than free).

We expect 'safe assets' to have a positive real return because we assume some combination of
1) Labor will always have a positive marginal product- hence 'time preference' based lending attracts a 'reward for thrift'.
However, if Labor has a negative marginal product- as may be the case for Technological, Sociological and Demographic reasons- there is no reward for thrift but just a transfer boosting aggregate demand which cancels out what would otherwise be a deflationary Pigou wealth effect on nominal assets.
It may be that rational 'proles' (i.e. those who only serve the State by reproducing themselves) expect lower real incomes in the future and turn the Permanent Income Hypothesis on its head by deciding to make merry before that long Winter sets in. In other words, Proletarian time preference isn't linked to an expectation of working more productively later on but just of enjoying one final carnival of consumption.
2) Positive Marginal efficiency of Capital obtains sooner or later along the Envelope curve.
However, this depends on Entrepreneurship's time preference. Suppose Uncertainty has increased for some exogenous reason- e.g some 'deep' Scientific paradigm shift which calls all current Technological Business Models into question- Entrepreneurs can either try to exploit this by making as much money for themselves (as 'snake oil' salesmen) now or else compete in a Darwinian 'regret minimizing' manner (e.g. investing in the twenty first century equivalent of Petroleum and the new type of products it makes possible) such that Society as a whole, and Financial Markets, move along a discernible 'golden path'. The problem here is that markets may capitalize their current income regardless of which path they choose.

Currently, it looks as though the Market trusts some 'dynamic' Corporations to sit on pools of cash and drive down the price of new start-ups which they eventually cannibalize to repeat the same process. However, the corollary is that other less apparently 'dynamic' Corporations which actually have positive marginal efficiency of capital are penalized such that even Self financed Capital Deepening reduces the share price.

Govts. ability to provide 'safe assets' was a feature of the Great Moderation which had a second life thanks to the disappearance of old fashioned cost-push inflationary pressures. However, Brexit- where stupid proles dependent of EU handouts cut off their noses to spite their faces- has raised a new bogeyman- not militant Unions, who could be briefly blinded by money illusion- but militant Stupidity and Chauvinism of the type that affected the Gadarene Swine.

I believe Cowen has covered the concept before, but "nonmarketable perpetual bonds" would provide the most stimulus. What are they? I suppose some would call it helicopter money: the Fed buys the bonds from the Treasury and the Treasury spends the money. In other words, there's no offsetting contraction, only expansion. As we continue the endless (and now pointless) debate about Lehman (see Professor Ball's critical paper), we ignore options that might make a difference today. Of course, the point of "nonmarketable perpetual bonds" is that they don't add to "safe assets": they are "nonmarketable" because the Fed wouldn't be allowed to resell them.

Personally I would be more impressed by the " shortage of safe assets " argument for low bond yields if inflation was at its target and the worlds central banks weren't determined to raise interest rates at the slightest provocation. What is going on is clear, monetary growth has been curtailed like never before and people are increasingly seeing bonds as a cash substitute. No one is worrying about a loss in value due to unexpected inflation anymore. Remember there is plenty of cash in banks and that is the same as a perpetual bond that pays no interest. So if you don't believe there will be inflation bonds and cash are substitutes except you might get some small interest on bonds and no bank charges.

If you want higher bond yields set level NGDP growth targets for your central bank.

In any event why worry about low bond prices? As a symptom of low growth they are bad, but taken in isolation low government borrowing costs are good.

In terms of low interest rates, I still don't see how the slowing down of population growth, especially working age population, is not the prime reason creating low rates as economies. Japan the last 25 years has pulled every single Friedman and Keynes lever and yet has had low rates the entire period. If their working age population has decreased from 86 million to 76 million, then show me where the economy needs lots of investment.

Indeed, the world is turning Japanese, due to Japanese demographics spreading throughout the world. Some say birthrates will rise again....I'm skeptical.

"I’m skeptical."

I know! Someday I'll convince you that there's evolutionary pressure to increase agreeableness...


All I'm saying is the current facts on the ground support my thesis, you offer far future hope and change. We'll see won't we.

No the Japanese have not "pulled every Friedman lever". They have not raised inflation by much because they don't want to, but the small increase the CB has allowed has driven their employment ratio to new highs. If they really want inflation forget fiscal boondoggles and focus on monetisation of their debt ( which by the way they are already doing).

Don't like 1 or 2, deficit still north of $500 billion and increasing. Don't like the intergenerational aspect (right Bernie?). Either one of these feels like Boomers continuing to let themselves off the hook.

3 sounds like self-insurance. Is this simply reviving the "ownership society" idea? To the extent rich countries can stomach this, sounds like the plan. To the extent individuals within rich countries can stomach this, they will prosper compared to those who can't or won't. The existence of a meaningful social insurance system for middle and lower income folks (Social Security and Medicare) provides, I think, an important and sometimes overlooked foundation here, but if we hope to pay these bills in 10 or 20 years' time, we can't be engaging in reckless deficit spending or tax cuts (1 and 2) in the meantime.

4. Thin

5. Well, the Fed has taken over $4 trillion out of play and onto their balance sheet. Should they start selling? Would that have the same effect? Maybe they could take all that money and use it to pay out the drop in reserves if they end IOR at the same time.

"Don’t like the intergenerational aspect "

I think the intergenerational argument against national debt has never been as weak as it is now. But has it ever worked? For example, can anyone make a reasonable argument that the Baby Boom generation was worse off because the Greatest Generation ran up huge debts fighting WWII (ignore the fact that WWII was necessary, whether defeating Japan and Germany was good or not that shouldn't change the economic fact that huge debt was incurred)?

IMO the intergenerational argument fails because all debt is balanced on the other side of the ledger as an asset. While Boomers are supposedly naughty today for building up gov't debt, the fact is consumption that could happen today is not happening because someone somewhere is opting to buy bonds the Treasury is issuing. Whether that's someone forgoing extra Starbucks runs every week to add to their 401K, a Saudi prince opting to not cover his yacht with gold, or whatever the fact is every effort to spin a morality tale out of it produces the exact opposite tale.

Yes. I understand your "governments aren't households" fig leaf. There are lots of ways for one generation to bequeath more to the next, or not. Winning a world war and then reducing THAT debt and overseeing a period of sharp growth in living standards compares pretty favorably with whatever story Boomers are telling themselves these days.

Yes winning WWII was necessary but that doesn't alter the fact that in your model it should be a huge cost.

Consider, say when you were young your house burnt down. Your parents then rebuilt that house from scratch by dipping into their savings, working day and night etc. Now they die and leave you the house. Guess the house burning down didn't hurt you did it?

But it did. If that house hadn't burned down you would have inherited it today AND the savings they had spent rebuilding it. So the house burning had a huge opportunity cost for you even if you don't happen to feel like you are suffering today because it happened.

Yet it is hard to make that case. If WWII didn't happen would the 1950's have arrived in the 40's? Would we have hit the moon in 59? Computers, iphones, Tesla all sooner? I don't think so. In fact I could see the post war decades being poorer if WWII never happened...

Yet your model says massive debt incurred in one generation means a huge burden put on the future generation to pay it off. That narrative doesn't stick with one of the largest debts incurred in recent history.

But how about the rest of history? Can you find any real examples of a previous generation imposing horrible debts on a future generation? If you can't how does your model account for this? Have all generations been remarkably good to future generations until the Baby Boomers? An inexplicable display of self-lessness in economics that normally assumes individuals act selfishly?

Here's an example: Greece.

The recent run up in debt has nothing to do with suffering through a Great Depression followed by a World War. Debt dropped dramatically after WWII. I'm sure your great uncle was calling for another 'austerity' recession in the late 1940s, like a good Keynesian.

Hmmm Greece....

From what I understand Greece essentially owes something like $100 on Friday. They want the EU to loan them $100 to make the $100 payment and then loan them another $10. After lots of back and forth Germany agrees to loan them $100 to make their loan payment provided they only get an additional $8 loan rather than $10.

While the Greeks of today aren't feeling great it hardly seems like they are taking on hardship to pay back loans made by Germans and others from back then.

"Winning a world war and then reducing THAT debt "

Not really. WWII ended in 1945 and we had about $259B in debt. By 1950 we worked that down to $252B. It never got any lower than that. To this very day we still owe about a quarter trillion from fighting WWII and it has never been paid off, it simply keeps rolling over except now it is a trivial share of our total economy.

Again what evidence do you have to show that the post WWII years were spent suffering to 'pay off' the debt incurred in WWII (not matter how necessary and noble that was)?

What happened to the deficit talk?

That's the opposite side of creating bonds.

OMG! Foul!

You can't talk about the deficit or the $19 trillion national debt or the $50 to $100 trillion in unfunded liabilities. They don't fit the narrative! Pay no attention to the man behind the green curtain!

Accounting is the language of economics and finance. The following may facilitate understanding what occurs when the Fed monetizes the deficit and, concomitantly, adds - deficit spending - to the national debt. The journal entry is as follows.

Debt (increase an asset): US Treasury Securities $xxx,xxx,xxx
Credit (increase a liability): US Treasury checking account $xxx,xxx,xxx

At the UST, the entry is the reverse. The Fed checking account (asset) increases and the outstanding UST Treasury securities - national debt (liability) increases.

The US then spends the money in its Fed checking account on . . .

Since end 2008, whatever they've been doing with $4 trillion in Fed QE's and $9 trillion in additional deficit spending, it ain't working.

Pro tip: if a Democrat starts talking about the debt or deficit, it means they think Trump might win.

All it took was a year of Trumpery to get the country to stop its incessant complaining about how this is the worst time ever for at least a minute. It won't last, but I'ma enjoy it for the time being.

SuperPro Tip: When Republicans begin talking about the need for safe assets you know there is a tax plan just around the corner that will create a massive deficit, as the Bush tax cut did, satisfying the demand for safe assets with more government bods. By the way, the Bush tax cut was preceded by comments by Alan Greenspan about how difficult it is to manage the economy with so little debt. (You can also look up Cheney's statement that deficits don't matter, said at the same time). But, you can really tell when the other star of Republican mythology arrives: the magic fairy of imagined growth from giving tax breaks to millionaires who will reinvest in the US (never mind that they can take their tax cut and buy foreign ETFs with it).

I feel like I didn't really make much sense of the whole safe asset thing until I read some John Cochrane on equity-financed banking. Saying we have (or don't have) a safe asset shortage one should immediately think of is that a demand or supply issue. Pretty much everyone talking about the issue mentions the supply of safe assets, when probably the issue is artificially boosting the demand for them.

green energy r&d and anti-cancer r&d seem to have a lot of headroom

Remember: insurance doesn't reduce uncertainty, it just moves around its impact. When I buy insurance, my risk goes down, and the insurance company's goes up. Since the insurance company is really a million shareholders, diversified, their risk goes up in aggregate less than mine falls.

When I trade in my GM stock for a government bond, my risk goes down, but the government's goes up. The government is really millions of taxpayers and expenditure-getters. To keep things simple, let's suppose the Treasury uses the money I pay to buy GM stock, so literally it's a trade. Now, if GM stock goes down, Treasury will need to raise taxes or reduce expenditure. This is a nondiversifiable risk for the taxpayers and beneficiaries. Hence, it is unclear if aggregate risk really falls.

The key to this is that the government's "stockholders" are involuntary (except via their votes). That's why government debt is safe for the holder. But it's also why it isn't really insurance from a social point of view.

True but buying insurance doesn't increase the risk either. If people start loading up on fire insurance then insurance company profits go up as the number of fires will remain the same (or even go down if people are so spooked by fire they start being more careful).

See p. 5-8 of my for info on cost-benefit analysis discount rates used by the US government, which is relevant to this discussion. The basic principle is to think about how much private investment is crowded out.

It depends on how the money is used, doesn't it. If debt is taken on to build long term assets--roads, infrastructure, investment in human capital (education)--there is an increase in productivity to pay for the debt. But, if, on the other hand, we spend $5 Trillion to buy a war in Iraq, I wonder what the offsetting future gain is.

It's how you spend the money. There are purchases you make today that pay out over the long term, and are paid back by those using or acquiring those assets over their lifetime.

It is just as important to think about how much public investment is crowded out as it is how much private investment is crowded out. Often they are complementary, not competitive.

Yes, it's important to look at the yield on public investment. In this MR post, though, the question is whether the government shoujld issue more debt just to supply the demand for safe assets, even if the government doesn't have any good investments to make.

Please show me the assumption you supplied: that the government doesn't have any good investments to make. I didn't see that in the post, although I did see the phrase "massive tax cuts", which, in itself is interesting: those with the most private assets demand a tax cut to get bonds to protect their accumulated private assets.

The poor as insurers of last resort. And, for them, the alternative use of money would have been better schools, lower costs of education, and other long term capital assets to their life. Which, of course, could be taxed later.

Economists have been debating for some years now whether a negative interest rate may be required for some stimulus effect.

Some years along the way, some genius comes along and asks "if that would work .... uh ... don't you think that might mean there is a shortage in safe assets?"

Well, this IS one of he primary means via which the US benefits from the status of the US dollar.

Whether or not overall it would be correct to emphasize this point is completely separate from the fact that it seems to have almost completely escaped interest for so long.

Perhaps the underlying problem is that too many people just want to sit and watch their money make money for them, while even trees and fish naturally yield far higher returns than much of what is on the market these days. Or ... perhaps the boomers DID manage a way to transfer all future advances into present prices as a means of transferring gains from youth to their generation, whether it was the intention or not. Fight Club had some ideas on how to deal with such a situation, but I'm pretty sure there's a much better way to address it ...

Also, since many investors can use opposing risks as a means of reducing portfolio risk with a much higher underlying risk of each asset, we should assume that the market will systematically underprice risk. No?

Comments for this post are closed