Not too long ago when the Australian government ran a budget surplus, the Howard administration announced a plan to stop govenrment borrowing. That was around 2003. The financial industry was unhappy with it and lobbied the government to abandon that plan, citing havoc it would cause in the Australian financial market. The lobby was succesful. The Australian government continued to borrow even in times of fiscal surplus.
The idea how absence of government bonds in the local market may cause havoc is simple. All interest rates are more or less dependent on interest rate of a risk-free asset. In most cases, a risk-free asset is a sovereign bond of a reputable government, which more often than not, members of the Organization of Economic Cooperation and Development, the OCED, which is a grouping of the most developed as well as the most influential economies in the world.
It is risk-free in a sense that these governments, and in this case, the Australian government, would not default on their obligation to service the debts. Given the certainty that it provides, others instruments are priced with the rates of sovereign bonds considered. In other words, government bonds provide benchmark interest rate for the financial industry to use for other purposes ranging from simple lending and saving activities to complex derivatives.
How much disturbance would it cause if a government ceases issuing bonds?
I am quite concerned with this question because as a libertarian of largely minarchist tradition, the argument provides a hurdle to smaller government.
Firstly, by connecting the centrality of sovereign bonds as risk-free asset to the health of the financial industry and the economy at large, it legitimizes government intervention in the market.
Secondly, in time of budget surplus, it prevents valuable resources from being used in other areas. Borrowing imposes cost and the cost is being borne for no productive spending at all. It is like Santa Claus throwing money to the streets, except that it is the taxpayers that ultimately pay for it. It is not so much an issue in time of deficit because such deficit spending is grounded on other rationale, regardless whether that rationale is acceptable or not.
Thirdly, borrowing in times when the government has little use for extra fund introduces an unnecessary opportunity cost. “Oh, extra money! Let us spend it”. After all, with interest charged on that idle money, surely there are better ways to utilize it. That involves reinvesting that borrowed money into investments that provide higher returns. Or funding new government programs that veer away from the role of a limited government. That is not a libertarian-friendly idea.
Returning to the question, how much disturbance or havoc?
I would argue not much since the market will adapt to a scenario without government bonds in the local market.
It is true that without government bond in the market, market players will not have a risk-free asset to base their pricing on, within local context. I am sure they will be able to substitute it with other assets locally however. It will not be risk-free but it is still high quality assets. That probably may cause cost of borrowing to go systematically up since the minimum interest rate in the market that forms the base of all pricings increases to correspond with greater risk faced by market participants. Nonetheless, the industry will find an alternative benchmark.
Furthermore, that alternative benchmark does not have to originate from the local market. Other governments do borrow and some of the most reputable governments, as far as fulfilling their debt obligation go, borrow massively. Save for foreign exchange rate fluctuation risk, there is no reason why the rate at which reputable foreign governments borrow cannot be the benchmark.
I suspect the argument against zero-debt made by the Australia financial industry players is about protecting their revenue rather than problem that it might cause to the market’s ability to price assets.
7 replies on “[2072] Of the market can live without government bonds”
[…] Now that we are staring at the possible end of the world — or rather, the possibility of US defaulting it payment — I think I am an optimist. Even if there would be a default, it would likely be temporary. Even if it would be permanent, I have made my position known: the market can live without government bonds. […]
Like I said, it’s opportunity cost. If a benchmark doesn’t act as the opportunity cost, then a benchmark is meaningless.
On prices as irrelevant and its yields that matter, well, price is the inverse of yield. So, it can be prices at discount, par or premium, depending on the yield.
Furthermore, currency denomination is only irrelevant in our case between time after purchase and time before sale or before maturity. Consider also at time of maturity. If the bonds pays in foreign currency and a person lives in Malaysia, surely, forex risk exists. The same is true for purchase.
Consider this:
A 1-year zero-coupon @ 5% yield of foreign govt bond. Local A holds this to maturity
vs.
A 1-year zero-coupon @ 5% yield of a local blue chip. Local B holds this to maturity. (Assume the blue chip is as riskless as the foreign govt bond so that we can focus on the opportunity cost that I mentioned)
Now, MYR weakens against that foreign currency associated with the foreign govt. MYR is much weaker at maturity.
At maturity, Local B is worse off than Local A.
With hindsight, Local B would demand for the change in exchange rate to be compensated if he were to remain indifferent between the two.
In fact, if the the scenario is to be played again and Local B has to choose the blue chip, Local B will want the uncertainty to be compensated upfront. This is the risk.
“By trade, rather, the consumer must have the ability to purchase the foreign government bonds.”
Why? This isn’t necessary for a benchmark to be effective.
“In absence of sovereign bond, the issuers of the next best bonds have to beat prices (denominated in foreign currency) faced by the consumers”
For benchmarking, prices are irrelevant – it’s the yield that matters. Which also implies that the currency denomination doesn’t matter either.
1. It’s not that the bonds need to be traded locally. By trade, rather, the consumer must have the ability to purchase the foreign government bonds. We may have contextualize the word trade differently here. Nevertheless, repeating my argument after the clarification, the forex risks comes because the consumers, presumably most of them, earn in MYR. In absence of sovereign bond, the issuers of the next best bonds have to beat prices (denominated in foreign currency) faced by the consumers. That’s the opportunity cost and that opportunity cost incorporates the forex risk.
3. Obviously it’s risky. Nevertheless, in absence of govt bond, it is the next best thing. It “flies in the face of market” simply because it hasn’t happened. If govt stopped borrowing, this is definitely one of the top possible scenarios.
On point 1, you miss my point – a govt debt benchmark establishes a floor price for other assets. Using a foreign government bond as the benchmark however doesn’t require that those bonds be traded locally as well – hence no forex implication. You can still establish pricing without domestic players trading in those bonds. There’s a precedent for this – many Euro-dollar issues are based on LIBOR, but the debt is denominated in USD not in GBP.
On point 3, that flies in the face of market realities. Unlike government bonds there are no mandated market-makers for corporate issues – it’s too risky. Second, there aren’t any corporates that issue standardised bonds with a wide enough range of tenures in sufficient volume to establish benchmarks on. In the absence of a liquid government bond market, risk premiums go up and issuance dies down – everybody goes back to bank loans because it would be cheaper.
Thanks for the comment.
1. If it doesn’t involve trading, I have two concerns. First, without trading, the issue of liquidity should not arise, contrary to your third point. Two, without trading, foreign government debts cease to be the opportunity cost and hence, it would stop being a benchmark.Therefore, especially based on my second concern, it has to be bought and sold as freely as local government bonds. The purchasers of any financial instruments will face the opportunity cost between local instruments and foreign government bonds. And because it has to be traded, then there has to be forex risk, unless there’s currency board in place. The purchasers face that forex risk and therefore, forex risk must be present.
2. This is agreeable though I think, premium incorporated in the exchange rate through the Fisher effect may somewhat answer that concern. But yea, I can agree with that.
3. This is true at the moment. But I think, if government stops borrowing, the market will take up this less traded but highly rated corporate bonds as the new benchmark and that would increase its liquidity.
A few comments:
1. Benchmarking against foreign goverment debt doesn’t attract forex risk – the benchmark provides a basis for pricing, there is no trading involved.
2. But using foreign benchmarks carry the risk of mispricing local risk, because of differences in monetary conditions and position in the business cycle.
3. On the other hand, it’s actually pretty hard to find a non-sovereign local bond benchmark for one very critical reason – lack of liquidity. About the only non-govt bonds that get traded are AAA-rated, and even then the volume is too low to establish confidence in the market price without govt debt benchmarks to follow.
That’s actually the key thing here – there actually are active secondary markets for govt debt across a wide range of maturities that can be referenced on a day to day basis, unlike other bonds which are typically held to maturity. And the price is market-determined, unlike close alternatives like the interbank rate, which can be directly influenced by central banks and which has too short a tenure anyway.