I read an article in the Financial Times earlier this week (September 23, 2023) -…
The Australian government released the business plan for NBN Co today which outlines the “cost-benefit” case for creating a monopoly wholesaler of fibre-based broadband services in Australia and investing some $A27 billion in public funds to create the network. The business case has been the focus of much political debate over the last year or so and as usual most of the debate has been conducted on a spurious basis – that is, the assumption is that the budget outlays proposed represent a “cost” to government and that by committing funds to this project the government is less able to “afford” other projects – presumably because there is some “budget balance outcome” that it cannot deviate from. Neither proposition is valid. While this blog has an Australian flavour the general economic principles apply to all national governments contemplating large-scale public infrastructure developments. The general point is that the provision of public infrastructure does not have to earn commercial returns.
The Government hasn’t helped its case by continually making statements that it is committed to getting the budget back into surplus as soon as possible without reference to what the other sectors (external and private domestic) might be intending. The budget outcome is not determined by government alone and by artificially imposing a constraint on itself it may well guarantee that the budget moves in the opposite direction to that intended.
By imposing these nonsensical fiscal rules on itself independent of the actual trajectory that private spending may take the Government is taking a high risk strategy. If private spending is weaker – and we have already seen that the exchange rate appreciation is reducing the capacity of the Government to achieve these fiscal targets – then the overall aspiration for a budget surplus may be impossible to realise.
It is far better for governments to acknowledge that the budget outcome is driven by a combination of discretionary government spending and taxation decisions in addition to the spending decisions taken by the non-government sector (including the state of net exports, household saving and private gross fixed capital formation).
Further, the decision to invest in a national broadband network (NBN) should not be taken on “financial” grounds. Rather the social costs and returns should be considered along with the availability of the real resources (and their competing uses) necessary to construct the infrastructure. I previously wrote about the NBN proposal in this blog – Free public broadband is required.
Anyway, I tried to access the 160-page business plan from NBN Co’s home page and got this message:
Not a good start I thought for a national broadband provider!
I finally was able to access the document at this – address – the PDF is 3 mgbs.
The document is very interesting in many ways not the least being the engineering detail involved. I found all those aspects fascinating. But the economics of the case is less than compelling and exposes the whole plan to its neo-liberal bias.
The goals of the NBN include:
- The network should be designed to provide an open access, wholesale only, national network.
- NBN Co should offer uniform national wholesale pricing over the network.
- The expected rate of return should, at a minimum, be in excess of current public debt rates.
So the first two goals (among others) are fine and reflect the fact that Australia has had an very poor record of broadband provision since the privatisation of Telstra (which effectively represented a natural monopoly in copper wire telecommunications).
The first two goals also recognise that Australia is a vast land mass with a highly dense population hugging the coastline and a much sparser population living inland but our national psyche says that all citizens should enjoy equal access to public infrastructure such as telecommunications, post and other essential services.
In practice, it doesn’t work out that way but the principle is ingrained in our public policy debates and large cross-subsidies to the regional and remote areas have long been the norm – particularly in the provision of telecommunication services. For years, the high volume long distance services between the capital cities has been milked (higher prices) to provide the resources to ensure that telephony is available to the low-volume areas in the regions.
I have long been a critic of the hidden cross-subsidy approach that successive governments adopted as part of their charter to ensure that public or partly-public owned enterprises fulfull what have been called community service objectives (CSOs). My argument has always been that if there are CSOs to be delivered then there should be an explicit recognition that real resources are being so diverted and that some indication of the extent of this “real” diversion should be publicly available.
Then the citizens can decide via the ballot box whether they want to collectively use these real services in this way with the full knowledge of what is involved.
However, the final goal listed above tells us that the neo-liberals are well and truly running this initiative and that this ideological bias is probably distorting the investment decision-making framework being used by the Government.
The NBN Co document (page 12) recognises “the need to cross-subsidise non-commercially viable market segments” and that “NBN Co will not be able to compete effectively with cherry pickers, who focus on commercially attractive areas only”. So legislative or economic (levies) protection will be required to prevent the cherry pickers from rendering the NBN uncompetitive.
On page 13, the plan says:
NBN Co has reflected the Government’s decision that the Company should implement an internal cross-subsidy to provide uniform national wholesale pricing over the network, from a PoI to a premises.
PoI = point of interconnection to the wholesale network.
On Page 100, the “pricing principles” are outlined. I found most of this discussion irrelevant although the comparisons between the proposed services (speeds, technologies) and existing capacity was interesting. The message is clear – we are being so poorly served at present. But why should NBN Co care about pricing in relation to existing commercial services?
The press conference today (when the Prime Minister released the report) was dominated by discussions of pricing. There is no reason why the NBN Co should seek a commercial return at all. This is a public good that is being provided.
The whole argument appears to be based on the Prime Minister’s claim that the “taxpayers will get all their funds back with interest”. The reality is that the taxpayer will not be putting any funds into the project. Please read my blog – Taxpayers do not fund anything.
By forcing NBN Co to deliver a near commercial rate of return, the Government is ensuring that the retail prices will be higher than is necessary for such an elemental service.
On Page 104, the NBN Co summarises the internal rate of return (IRR) that is expects.
The response from News Limited via the national daily – The Australian (December 20, 2010) was that – NBN return to be lower than in a commercial world, business plan shows – to which I said so what?
The Australian article said:
THE National Broadband Network will deliver an internal rate of return of 7 per cent … NBN Co boss Mike Quigley said the internal rate of return for the business case was lower than might be expected in a commercial business plan … It will also be profitable, meaning taxpayers’ investment in the NBN “will be returned with interest”, while uniform wholesale prices will also be achieved.
There are several different cost-benefit indicators that can be used when determining capital budgeting and assessing different uses of such capital.
The Internal Rate of Return is one such measure and is often referred to as the “the discounted cash flow rate of return (DCFROR) or simply the rate of return (ROR)”.
For those not familiar with discounted cash flow analysis, we recognise that a large-scale project involves revenue and outlays over multi-periods and to assess the return on such a project over some “lifetime” we have to devise a way to compare these temporal flows on a consistent basis.
The Present Value concept recognises that a dollar today is worth more than a dollar tomorrow because it can be invested and interest gained. So revenue or costs in future periods cannot be readily compared with revenue and costs now.
To overcome this temporal issue, we define a present value by “discounting” future cash flows to take into account the “time value of money” and “investment risk”. We thus use some interest rate (discount rate) to bring all future revenue/outlays back to a present day value. You can look up formulas to see how this is done if you are interested.
The net present value (NPV) is just the sum of the revenues (in present value terms) minus the sum of the outlays (also in present value terms). The NPV indicates the value on a long-term project and if the NPV > 0 then the project adds value to the firm.
There is a huge debate about the short-comings of NPVs as a guide to capital expenditure. I will leave it to your curiosity to pursue the literature further should you be interested.
In the context of this blog, one such issue is that NPVs do not tell you the profit in percentage terms of investing in a specific project. They merely indicate the value of an investment. They do not provide information about the efficiency or yield of the investment. In this context, the use of the internal rate of return is indicated in the literature.
Imagine you want to know the break-even discount rate where the present value of all outlays equals the present value of all revenue. That discount rate is the IRR. The higher the IRR, the more desirable the project.
The finance literature claims that a firm should undertake all projects where the IRR > cost of capital. It is for this reason that the NBN Co’s business plan makes it clear the the IRR (7 per cent) exceeds the long-term bond rate (the yield associated with the issuance of long-term public bonds). But in the biased parlance of the government, the long-term bond rate is what they think is the “cost of raising funds”.
The literature on discounted cash flow analysis is full of warnings about the pitfalls in using IRR measures to guide capital budgeting. One major qualification relates to interim cash flows that a project might or might not deliver over the life of the project. For example, see this this article.
However, the concept – the cost of raising funds – has no application to a sovereign government which is not revenue-constrained.
To make headway we need to take the National Broadband Plan out of the “free market rhetoric” and see it as an essential public good that is best constructed as a “natural monopoly” by the national government which faces no financing constraints. Then you have a different perspective altogether and most of the debate that is going on at present falls away as irrelevant and its ideological basis becomes immediately obvious.
First, why should the provision of the optic fibre network be evaluated in commercial value (read: what will make profits for a private corporation)? In saying that I am not considering so-called value-adding services that might use the network. Just consider the network itself – the optic fibre into our houses.
Why should that not be seen as essential public infrastructure? In that context, it is immediately obvious that it becomes the legitimate responsibility of the Federal government to provide it to advance public purpose? Clearly if we ignore the neo-liberal biases that have derailed the global economy after years of eliminating public provision of essential infrastructure the benefits of having a universal and single infrastructure based on best-practice are obvious.
Second, what does it actually cost? There is very little discussion of this in the NBN Co’s business plan. How many workers are going to be employed? What skills mixes will be required? What training is NBN going to provide to ensure the huge pool of idle labour currently available will benefit from the infrastructure development? And questions like this that relate to the real resource implications of the project …
The only reference to employment appears on Page 153 of the Plan, where we read that:
Under the Department of Education, Employment and Workplace Relations, these principles support the creation of quality jobs by ensuring that NBN Co procurement decisions are consistent with the Fair Work Act and its aims, including promoting fair, cooperative and productive workplaces.
Okay, that is a good principle – the NBN is not going to undermine employment conditions.
Answer: there is nothing in the Business Plan about these issues. So the rate of return analysis is in my view irrelevant. I want to know what real resources are going to be deployed and what alternative uses these resources may have been occupied in. Then we can consider the true costs against the social benefits forthcoming.
However, the general principle is that if the real goods and services that are required to construct it are available then the Federal government will be able to purchase them using deficit spending. No debt issuance is required. The only reason that debt might be issued coincidently with the spending incurred to construct the network is if the Federal government considered there was too much liquidity in the private economy which was thwarting its overall economic management (most notably monetary policy).
It is highly likely that the results of the current economic downturn will be a protracted period of under-utilised capacity, particularly labour and so net government spending (rising deficits) will be extremely beneficial to the economy.
So arguments about how the government will pay for the broadband network (and the “costs”) are largely irrelevant. It will pay for it like it pays for anything else by signing cheques and crediting private bank accounts. Then it will monitor the general state of the economy and it might increase taxes or issue bonds if they want private purchasing power to be less.
Of-course, it might also decrease taxes and buy back bonds if it considers more private purchasing power is required. You can see the point. These liquidity management issues are not related in any causal way with the decision by the Federal government to net spend in the AUD to build the broadband network.
Further, once we clear the neo-liberal baggage surrounding “financing issues” then we can further dismiss the arguments about the broadband network being doomed to fail because the high access fees that will be necessary to recoup commercial returns will choke off household demand. These arguments are just nonsensical and rely on us believing that the only value that matters are commercial (privatised) returns.
Why are we being manipulated by all and sundry into constructing this infrastructure solely in terms of a market-based (privatised) commercial project?
In 2003 I wrote this article – Divisions over public debt – for the self-styled progressive organisation the Evatt Foundation, which was set up to honour H.V. Evatt a prominent Labor politician during the 1940s and 1950s and which has been largely funded by state and federal labor governments.
The article noted that
The recent Commonwealth Treasury Debt Management Review highlights the damaging divide among ‘progressive economists’ in Australia when it comes to macroeconomic analysis. This fracture is greater than the things that bind us and helps the neo-liberal orthodoxy to defend their appalling economic record.
This was in reference to a public enquiry set up by the Federal Government in 2002 to address concerns expressed by the Sydney Futures Exchange and other beneficiaries of corporate welfare in the form of public debt. The conservative government had been running consistent budget surpluses and retiring public debt to the point that the bond markets were became too thin and this threatened the welfare received by the bond traders in the form of a ready supply of risk-free annuities that the government bonds provided.
At the time, two submissions were made to the Treasury Review by progressive economists. The Evatt Foundation submission, based on papers by Tony Aspromourgos and Frank Stilwell, was the polar opposite to that submitted by the Centre of Full Employment and Equity (CofFEE) (written by myself and Warren Mosler).
The other review submissions were largely self-serving pleadings for continued government assistance from the major players in the Commonwealth Government Securities (CGS) market. Why would the Evatt Submission reach the same conclusion as the financial markets?
I noted that the:
The equations in the Evatt Submission are derived from the GBC model. Stilwell also operates within this framework, saying that “Governments can finance their expenditure either by taxes or by borrowing … [and with political constraints on taxation] … reduced government borrowing means lower capital expenditure. So the deterioration in public infrastructure – in the quality of ‘public goods’ in general – is a direct consequence of the commitment to debt reduction”.
Immediately, the argument then becomes one of ‘why isn’t there more debt issuance’ as opposed to ‘why aren’t there budget deficits’. The decline in the quantity (quality) of public goods is a direct consequence of inadequate levels of government spending and has nothing at all to do with the extent of CGS issuance. To link the two essentially independent functions is to constrain the macroeconomic analysis within an orthodox paradigm.
The reality is that the Australian government like all sovereign governments does not have to finance its spending either by raising taxation revenue or borrowing. A sovereign government is never revenue constrained because it is the monopoly issuer of the currency.
This has to become a fundamental postulate of progressive economists or else they will always be locked into the nonsensical mainstream logic that is derived when you interpret the government budget constraint (GBC) as representing an a priori financial constraint. The only sensible (and boring) interpretation of the GBC is that is is an ex post accounting statement of the stock-flow relationships that pertain to government spending, taxation and debt-issuance.
The GBC never tells us that there is a binding financial constraint on government spending. The fact that governments issue debt and continually talk about taxpayers’ funds just reflects their obedience to the dominant neo-liberal paradigm rather than saying anything intrinsic about the operational possibilities of the monetary system it operates.
The ‘financing’ trichotomy in the GBC (taxes, printing money, borrowing) is not a robust description of the way in which government spending and financial markets work. The government spends by crediting an account at a Reserve Bank of Australia (RBA) member bank.
This process cannot be revenue constrained (government cheques don’t bounce). Governments don’t spend by ‘printing money’. Taxation consists of debiting an account at a RBA member bank. While the funds debited are ‘accounted for’ they don’t actually ‘go anywhere’ nor ‘accumulate’.
Please read my blogs – Taxpayers do not fund anything and On voluntary constraints that undermine public purpose – for more discussion on this point.
Stillwell also said that:
Of course, government borrowing involves a cost – the interest payments on government bonds, for example … But whether the interest constitutes a ‘burden’ depends upon how it compares with the social benefits arising from the government spending. As in the case of the carpenter’s personal debt, the interest payments are not a net burden if the future income-generating capacity is enhanced.
This is the old Keynesian defence of the unnecessary and is almost apologetic in nature. It is a sort of ‘well if it is good for the household then it must be good for government’. In doing so we are left with the false neo-liberal analogy between the private agent (user of the currency) and the government (issuer of the currency).
The factors that lead a private investor to borrow are varied, but all of them are irrelevant for the government spending decision. If there is a social ‘rate of return’ for a particular item of government spending then it will accrue whether there is debt issued or not. To say that some public debt is wise and some is unwise, as the Evatt Submission does, is to fundamentally misconstrue the purpose of the debt issuance – which is to drain excess liquidity to allow monetary policy to work.
I concluded that article by saying that once we accept that government spending is not financially constrained and appreciate that public debt issuance is about monetary policy we can see that progressive talk about an optimal debt ratio is both irrelevant to achieving full employment and damaging to the progressive cause.
An economy can easily sustain zero public debt issuance without aberrant economic consequences. The important point that must bind progressive economists is that the government must net spend to sustain aggregate demand levels consistent with a full employment objective.
It is clear that more discussion has to occur between progressive macroeconomists to search for ways in which our differences can be resolved. In general, I believe the answer lies in rejecting the orthodox paradigm outright and gaining a sharper appreciation of how financial markets actually work. It is very different to the way that orthodox macroeconomics textbooks pretend they work.
A further erroneous claim is that by using public outlays to provide a better infrastructure for the middle-class to watch high-speed IPTV the Government is reducing its scope to reduce hospital waiting queues or improve public transport systems or ensure public schools are first-class.
This claim stems from the underlying bias that the budget should be balanced. Take the following diagram which depicts two alternative uses of public service provision.
So the budget will be balanced if (ex post) tax revenues are equal to public spending. Assume the government has only two spending options – national broadband and public hospitals (the latter could be considered all other public options).
The downward sloping line depicts all combinations of spending on the two options that would be consistent with a balanced budget. Point A is one such mix. The more spending the national broadband network absorbs the lower will be the provision of public hospital services along this line. The extremes are all broadband or all hospitals. This sloping line is just a “balanced budget possibilities” frontier.
The policy debate is continually being conducted as if this downward sloping line is a real constraint whereas, in fact, it is just a voluntary fiction imposed on governments by political forces of the right.
The point is that the sloping line is an artificial (ideological or political) constraint and thus has no economic or financial basis.
The Australian government has actually imposed an even harsher voluntary constraint (the red line) because it is trying to pursue budget surpluses even though there is more than 12.5 per cent of available labour resources idle.
Further, while an increased provision of both hospital care and the national broadband network is entirely dependent on the availability of real resources and has nothing to do with financial capacity, the point ignored by the neo-liberals is that such an increased provision can actually ensure a wider set of possibilities for the private sector.
The NBN Co Business Plan makes a convincing case that the private sector will enjoy enhanced benefits as a result of the higher quality broadband infrastructure. So in fact the social returns (taking into account all the externalities) are likely to be very significant.
The whole discussion of rates of return is misleading. In a future blog I will address what criteria should guide public sector capital infrastructure provision.
We need a fundamental re-think about how the role and status of the national broadband network. It should not be considered in commercial (private market) terms. Using commercial logic is likely to lead to under provision of the infrastructure and excessive pricing of its services. The social gains alone are likely to be significant.
Further, it is clear that building a first-rate broadband infrastructure will also increase employment growth (the sheer size of the investment will do this) and stimulate all sorts of new private businesses to innovate in the areas noted in the previous paragraph. So the private sector spins off the public infrastructure but not in a way that compromises the fundamental efficiency and equity aims of the exercise.
Whether you want private retail providers to exist is another question that I might address another day. But if there are significant social returns (not accounted for in the NBN Co’s business plan then the “rate of return” is likely to be much higher than they have estimated and therefore the wholesale prices on offer could be much lower – or even zero.
Whatever, the provision of public infrastructure does not have to earn commercial returns.
That is enough for today!