NEWS & ANALYSIS

Lessons from the Gauteng e-toll controversy - Jeremy Cronin

SACP deputy GS says Helen Zille and Jack Bloom's arguments are wrong-headed

Red Alert: The e-toll saga - ideological confusions and strange bedfellows

The e-toll debate has often been presented in the media in crude "for or against" terms. As a result the underlying socio-economic (and therefore class) issues have been seriously obscured.  There was even a brief moment last week when the e-tolls had some COSATU and DA leaders flirting with each other. Differences on labour brokers spoiled the party and provided a handy alibi for not pursuing the budding romance. But how could this fleeting affair have been possible in the first place?

To get to the bottom of this, a number of conceptual and factual clarifications are required.

The "user-pay principle"

I have heard some bureaucrats speak of the so-called "user-pay principle" as if it were the Eleventh Commandment itself, a universal principle that government must and will enforce. On the other hand, some comrades in COSATU present the user-pay approach as an inherent evil, as the "commodification" of a public good, applying "free market principles" to public roads.

Both positions are equally wrong or, at the very least, dogmatically one-sided. Even in a socialist economy, like Cuba`s, there are charges for certain categories of public goods - household electricity, for instance. But in Cuba (and indeed in SA) household electricity tariffs are not primarily based on a profit-making principle, or even on a rigid cost-recovery basis - although partial cost-recovery is part of the objective. Subsidized household electricity has been "de-commodified" in Cuba through applying a non-capitalist pricing-mechanism that is primarily focused on demand-regulation - to encourage households to use electricity wisely and efficiently. Nor are the Cubans satisfied with a pure, economistic price-mechanism alone. The electricity price-mechanism is used alongside extensive popular mobilization and campaigning. The UJC (the Cuban Young Communist League), in particular, has been active in a major door-to-door campaign to help households and work-places replace energy-guzzling electrical appliances. 

Roads, like household electricity, are (or should be) a public good. Here in South Africa, the construction and maintenance of our 535,000kms of proclaimed roads are, in principle, overwhelmingly funded directly out of the public purse. However, as we all know, the funding is massively inadequate and there is a huge backlog on maintenance that stretches back several decades.  A public agency, the SA National Roads Agency Ltd. (SANRAL), looks after slightly more than 16,000 kms of roads designated as "national" roads, of which 3,120kms, i.e. less than one-fifth, is tolled. Our national road network is generally in good shape - particularly, but not exclusively, the tolled portions.

But there is bad news for those bureaucrats who believe that the direct-user-pay principle is the silver bullet to solve all infrastructure funding problems. Apart from public resistance, it is simply not feasible to toll the majority of our national road network (let alone the rest of our roads). Sufficient volumes of traffic are required to cover the costs of a public or private toll-operator for ongoing maintenance responsibilities, and for the administrative costs of tolling itself, including the wages of the workers in the system.

But what are the possible arguments FOR tolling?

There are several potential advantages for tolling in APPROPRIATE circumstances:

First, tolling can enable road maintenance and construction off-budget - relieving the budget for other priorities - like health, education, public transport infrastructure, or our seriously underfunded (and inherently untollable) rural roads.

Secondly, a tolling project enables a public entity like SANRAL to borrow money on the markets up-front (using the collateral of its national road network). This enables SANRAL to run a major, multi-year construction project more effectively and with greater confidence than with annual budget allocations, where there are always funding uncertainties about the outer-years of a multi-year project. Even with our three-year Medium Term Expenditure Framework process, the second and third year allocations are not cast in stone.

Thirdly, tolling can - if WISELY and APPROPRIATELY used - introduce a degree of equity into road-use.  A single-axle of a large road freight truck causes 40,000 times more damage to a road surface than the average light vehicle! While road freight operators pay tolls (where they exist) and they pay for licenses and the standard fuel levies, they are paying a pathetically small amount relative to the public infrastructure they are using and consuming. By contrast, Transnet Freight Rail has to pay for the construction and maintenance of its own rail infrastructure.

This is one of the reasons why our publicly-owned rail operator battles to compete with privately-owned road freight operators, and why rail has lost very significant market share since the late 1970s when road-freight was deregulated. No-one is suggesting that we charge trucks tens of thousands times more than light vehicles. The impact of doing that on our economy (for example on food prices) would be unsustainable.  However, a different tolling rate for different vehicle types enables a degree of redress into this highly inequitable reality. Besides, the road freight operators actually save significantly in terms of diminished vehicle wear and tear on well constructed and well maintained roads. They should, in principle, have a larger interest in contributing to better roads.

Fourthly, tolling (and particularly electronic tolling) can also be used to achieve better developmental and sustainability outcomes as part of what is referred to as travel demand management (TDM).  In the case of the Gauteng Freeway Infrastructure Project (GFIP), time-of-day concessions on trucks, to encourage trucks to move at off-peak periods, might contribute to improved road safety and less congestion. Less congestion and therefore fewer vehicles crawling along gridlocked highways means, in turn, less petrol consumption and fewer green house gas and particulate emissions.

Fifthly, IF THERE ARE EFFECTIVE PUBLIC TRANSPORT NETWORKS IN PLACE, e-tolling infrastructure can be used for "congestion charging". This is being used very successfully in many cities, London being an excellent example. Electronic tolls imposed on private vehicles entering the inner perimeter of a city, and other measures like hefty charges on CBD parking severely discourage private vehicles entering into city centres to the benefit of public transport, pedestrian safety, the clearing of unsightly parking lots, and cleaner air and healthier precincts in formerly congested and polluted CBDs.

In short, in many respects, when APPROPRIATELY used, tolling (and other user-pay measures for public goods) could be a public sector intervention to correct "market failure" and structural distortions. Tolling can be the very opposite of a market-based commodification of public space. And this is the underlying class basis for a DA ideologue like Jack Bloom being so apoplectic about the Gauteng e-tolls. His argument is essentially a market-based argument - "I`ve paid, now give me"; "We the tax-payers (as if the DA`s largely white, middle class constituency were the only ones who paid taxes) have already paid for these roads - why should we have to pay twice?"

Note that Bloom and the rest of the DA are, by and large, not criticizing the appropriateness of a multi-billion rand spend on infrastructure that overwhelmingly (although not exclusively) services the needs of the relatively affluent in the richest corner of a highly unequal country. Bloom's beef is that he doesn`t want to pay for his lifestyle. 

So does this mean that COSATU and other progressive formations are wrong to be raising serious concerns about GFIP and e-tolling? The answer is a resounding: No. However, the basis on which COSATU has raised many of its concerns, and the solutions offered (basically: "dismantle the e-toll gantries") don`t helpfully take us forward either.

Reasons to be extremely cautious about the "user-pay principle"

In the preceding section we have raised several reasons why, when used appropriately, tolling can be a developmental intervention. However, precisely many of these potential advantages are also potential pitfalls.

In particular, given the advantages of borrowing capital upfront for a major multi-year road construction project against the promise of repayment through tolling, there is a danger of unduly favouring and prioritizing "tollable" projects to the detriment of other projects that might be more important from a developmental perspective.  "Tollable" projects are likely to be those in which more affluent users have a significant presence.  In the view of the SACP this is one of the key shortcomings of the GFIP e-toll process.

Clearly, it is not only affluent private car users who are on the GFIP network. Given the poor state of public transport and the continued geographical displacement of the working class into distant, peripheral dormitory townships, surveys have shown that in South Africa households tend to acquire a car at a very low income threshold. Basically, these are working class and lower middle class families who cannot really afford to run a car, but who equally, because of their geographical marginalization, cannot afford not to have a car. These are families who will be most severely hit by even marginal increases in the cost of running a car.

Notwithstanding this observation, we need to bear in mind that around 60% of Gauteng households (the poorest) have no access to a car whatsoever. Moreover, a recent vehicle count on the 180kms of the GFIP that will be tolled from next month suggests that only 2% of vehicles are public transport vehicles (minibuses and buses). This tells us that, at public expense, we have now spent R20bn on widening freeways in Gauteng that are largely (but not exclusively) used by relatively affluent private car users. Leaving aside tolling or not, was this the right infrastructure priority on which to spend R20bn? Should we not have spent it on mass transit public transport, for instance? (Much the same question needs to be asked about the R27bn spent on the elite-oriented Gautrain.)

So how did we get here?

GFIP e-tolling was not a clandestine project. Both GFIP and the Gautrain were widely and breathlessly heralded back in 2008 as a "new era" for Gauteng, turning the province into a "first world", "world class city-region".   Newspapers like The Star, which are now opportunistically jumping onto the anti-toll bandwagon, ran glowing editorials. Nor were the likely toll charges a mystery. SANRAL back in 2008 published an indicative 50c a kilometer estimate.

The major motivation for the project in the feasibility study produced by the UCT Graduate School of Business was that adding extra lanes onto Gauteng freeways would relieve the serious congestion challenges on the network. Nowhere in that study is any reference whatsoever made to what has become a commonplace in international scientific traffic research - namely, that expanding, improving and extending freeways generally tends, within a matter of years, to worsen congestion! This is what is known in the literature as "induced demand".

More freeway space attracts more vehicles and an expanding network encourages more urban sprawl as property speculators build more out-of-town shopping malls, golf-estates, and Midrand-style, Tuscan housing dormitory estates for the middle classes. While tolling might strand poorer commuters, those currently most responsible for the congestion, the more affluent, in their petro-guzzling 4X4s will hardly be fazed by the R550 monthly toll cap. They will continue to spur more dysfunctional urban sprawl by settling in increasing numbers in their dispersed, pseudo-country estates linked to work and leisure by expensive freeways.

Internationally it is widely understood that congestion is best addressed through other means - a vastly improved public transport network; mixed-income and mixed-use corridor densification along public transport routes; more effective public regulation of land use, including a much tougher grip on property speculators; congestion charging and other travel demand measures (like tough parking restrictions and parking charges in CBDs); and by shifting more freight off road onto rail. Spending R20bn on measures of these kinds would have been much better advised to combat congestion in Gauteng AND such measures would have multiple transformative and developmental outcomes as well.

These are some of the self-critical lessons we need to learn from the e-toll crisis.

But what practically can we do about the R20bn debt?

"Don`t pay it", appears to be the advice from some quarters. Unfortunately, that is simply not viable. Failure to honour the debt will make SANRAL (and the state - the guarantor - via the public purse) immediately liable, not just for this debt, but for all of SANRAL`s borrowings (around R37bn). This will be R37bn coming out of the budget and away from much greater spending priorities. We also now know that a default on this debt will have a serious impact on the Government Employees Pension Fund, and possibly on other private pension funds invested in SANRAL bonds.

"Increase the fuel levy to pay for GFIP" - is the argument coming from the DA and various sectoral interests (like road freight and the AA). This would mean that the whole of South Africa would be paying a significantly increased fuel levy to fund multi-billion rand infrastructure in Gauteng. The Jack Bloom argument that "we have already paid for the infrastructure through the fuel levy" is as wrong-headed as Helen Zille`s argument that the fuel levy should be ring-fenced for roads.  In fact, if there had been ring-fencing in the 2009/10 budget (the latest year for which I have a clear breakdown) roads would have received a R20bn allocation, whereas R26bn was actually allocated to national, provincial and municipal roads in that budget year.    

If we could rewind the clock, I am quite sure that the present administration would not give the go-ahead to an e-tolling project of this kind. Unfortunately the horse has bolted on at least GFIP Phase A1 (some 180kms of a planned more than 500kms of tolling on Gauteng freeways). The Gauteng Provincial government , supported by the Department of Transport, has put any further expansions on ice. Likewise, Minister of Transport, Sbu Ndebele has put on hold other SANRAL tolling projects in other parts of the country.

These are wise steps, and they lay the basis for a broad, and hopefully more unifying Tripartite Alliance discussion not so much on how to pay for freeways - but more importantly on what are our key infrastructure spending priorities.

Jeremy Cronin is deputy General Secretary of the SACP. This article first appeared in the Party's online journal Umsebenzi Online.

Click here to sign up to receive our free daily headline email newsletter