Chapter 8 - Investment

Provisions of AUSFTA relating to investment

8.1 Chapter 11 of the AUSFTA sets out the obligations of the parties in relation to investment.

8.2 In Australian public discourse, the term 'investment' is often used in a somewhat narrow sense, to describe investment in financial products or real property. In the AUSFTA, however, 'investment' is given its technically accurate meaning, and therefore includes almost any activity which involves the commitment of resources in return for reward, with the acceptance of risk. The DFAT Guide to the Agreement offers a series of examples of investments, including the following:

8.3 It can be seen from these provisions that the investment provisions capture an extremely broad range of economic activities. They are consequently extremely important in the context of the agreement as a whole.

Requirements under AUSFTA

8.4 Under the AUSFTA, the Parties agree to provide investors from the other Party either national treatment (that is, the same treatment afforded to domestic investors) or most-favoured-nation treatment, whichever is most advantageous to the investor. It also contains a number of provisions designed to reduce sovereign and other policy-related risks associated with investors from each Party investing in the economy of the other Party.

8.5 Specifically, the Parties agree to:

Reservations

8.6 Not all investment falls under the AUSFTA. Annex I and Annex II of the Agreement contain reservations allowing Parties to maintain existing non-conforming measures. Both Australia and the United States have included measures relative to investment in Annexes I and II. Any matter not mentioned in Annex I or Annex II is subject to the AUSFTA by default. Such investments are referred to in the Agreement as 'covered investments'.

8.7 Australia's reservations include:

8.8 The United States' reservations include:

8.9 Submissions and evidence raised a number of issues of concern in relation to Chapter 11 of the AUSFTA. These were foreshadowed briefly in the Committee's interim report[596] and will be discussed in greater detail below.

Overall impact on investment

8.10 The central concern is clearly to determine the impact the AUSFTA will have on investment in Australia. On the one hand, the Agreement may encourage additional investment in Australia by United States investors. On the other hand, it may encourage Australians to invest in the United States where previously they may have invested at home. Inevitably, of course, both of these situations will occur, and a great deal of energy has been spent during public debates on the AUSFTA in recent months trying to anticipate what the net effect on investment is likely to be.

8.11 The Department of Foreign Affairs and Trade cited investment as one of the most significant benefits of the AUSFTA:

It is in the area of investment that the gains from the Agreement will perhaps be most significant over time. Already the United States supplies nearly thirty per cent of Australia’s foreign investment, more than any other economy. Australia ranks 12th among destinations for US direct investment abroad. The United States is the biggest destination- 43 per cent- for Australia’s foreign direct investment (FDI), and Australia is the 10th largest foreign owner of US assets.

The Agreement will enhance Australia’s attractiveness as a destination for US investment as it puts in place legal guarantees and other measures that provide greater certainty for investors. The negotiation of the Agreement has already made Australia a greater focus of US investor and media attention, and this will continue during the US domestic approval process and beyond.[597]

8.12 The official analysis of the AUSFTA by the United States International Trade Commission does not attach the same significance to the investment aspects of the Agreement, offering a more subdued assessment:

The FTA will add transparency to the investment regimes of the united States and Australia, but is not expected to generate significant amounts of new investment between the two countries, as the investment environment in each is already substantially open.[598]

8.13 These more subdued claims are also reflected in the government's own Regulation Impact Statement:

Similarly, Australia's commitments under the Agreement with regard to screening of foreign investment is unlikely to have a major impact on US investment in Australia? [599]

8.14 Attempts to quantify, or to quantitatively verify these alleged gains have given rise to significant debate and various alternative analyses. The Department of Foreign Affairs and Trade commissioned the Centre for International Economics to assess the economic impact of the AUSFTA, which it did using the G-Cubed and Global Trade Analysis Project (GTAP) models. With respect to investment, the CIE made the following observation:

Lowering transaction costs, strengthening the security of the investment framework and highlighting the openness of Australia's foreign investment regime in non-sensitive sectors have the potential to reduce a proportion of the risk-related cost of capital in Australia (which reflects investor uncertainty and transaction costs). It is therefore likely that, whatever the impact on actual investment flows, the overall impact of the investment provisions in AUSFTA will be positive.

A difficult question is how positive is the result likely to be and how will the gains be distributed, particularly in more illiquid segments of the market?[600]

8.15 The CIE endeavours to quantify this benefit by indicating the likely fall in equity risk premiums on the implementation of the FTA. The term "equity risk premium" is a useful measure of the cost of capital. A reduction in the restrictions and risks associated with investing in a particular economy leads to a reduction in the risk premium, and therefore to less expensive access to capital for investee enterprises. The CIE, in a process outlined on p. 34 of its report, arrives at a conclusion that under the AUSFTA the risk equity premium would fall by 5 basis points (or .05 percentage points). It should be noted, though, that their analysis delivered an outcome of 10 basis points, which the CIE then halved in order to deliver a conservative result. It should also be noted that this figure does not appear to emerge from either G-Cubed or GTAP, but rather from a 'pragmatic'[601] series of calculations outlined in the report.

8.16 In a newspaper article in May 2004, Professor John Quiggin made the following observation critical of the CIE finding:

8.17 In an assessment of the CIE findings for this Committee, Dr Philippa Dee argued against the CIE's link between the AUSFTA changes to Foreign Investment Review Board (FIRB) assessment of investments (discussed further below) and the equity risk premium in Australia:

The equity risk premium is a concept that captures the effects of events that happen ex post, after an investment is made, that reduce or eliminate the expected returns on that investment. A negative ruling does not put at risk the entire amount that would have been invested. The potential investor still has their uninvested capital that they can put elsewhere-

There is no doubt that events that affect a country's equity risk premium can have a powerful effect on investment inflows, and hence on output and consumption levels in a country- A key factor likely to account for Australia's apparent equity risk premium is that we have a commodity-driven currency, so that the repatriated value of an investment in Australian manufacturing can be greatly affected 'after the event' by the price Australia gets for its wheat or coal.[603]

8.18 If Professor Quiggin and Dr. Dee are correct, and the AUSFTA does not result in a significant reduction of equity risk premiums in Australia, this in turn is likely to result in a dramatic reduction in the forecast economic benefits from the AUSFTA. DFAT, however, rejects Dr. Dee's assessment of the impact of the changes to FIRB review thresholds:

The CIE found substantial gains to Australia from the liberalisation of FIRB restrictions under AUSFTA. These findings are rejected by Dr. Dee on the ground that FIRB liberalisation will not reduce the risk premium on investment in Australia. But there is evidence that complying with its provisions is seen by investors as onerous. In addition, other provisions of AUSFTA will improve the investment climate and create added certainty for investors. The CIE's modelling assumes a very small reduction in the risk premium of only 5 basis points, and in sensitivity analysis, this is reduced to 2 basis points.[604]

8.19 In evidence the Department of the Treasury was reluctant to endorse the CIE's expectation of a 5 basis point reduction, but noted that in its view the precise number was less important than the policy message it signalled:

Investment in particular is a very difficult activity to model, so the modellers have come up with something that is logically consistent and fits with what we call mainstream theory, but we also need to keep in mind that this is giving us an indication.[605]

Dynamic productivity gains and economy-wide benefits

8.20 As the Committee noted in its interim report, the impact of Chapter 11 of the AUSFTA on the overall economic outcomes for Australia will depend substantially on the impact of dynamic productivity gains (or losses) arising from the AUSFTA. The CIE Report outlines the concept in the following terms:

It is widely accepted that trade liberalisation allows countries to move resources to more valuable sectors of the economy and consequently brings about allocative efficiency gains. However, trade reform does more than simply shift resources. When trade barriers are reduced on an industry, competition increases. Increased competition can change the behaviour of firms in that industry, encouraging businesses to use better technology and business practices either through innovation or quicker adoption of new ideas. Improvements to efficiency due to improved work practices (as opposed to resource re-allocation) are referred to as 'dynamic productivity gains'.[606]

8.21 Following World Bank research, the CIE identifies four sources of dynamic gains, two of which are relevant to investment. The CIE describes those two as follows:

8.22 Because of the multitude of factors which can influence the dynamic productivity changes associated with a policy measure, and because consideration of dynamic productivity changes is a relatively new science, such gains are extremely difficult to quantify with any accuracy. The CIE has dealt with this difficulty by developing, based on a series of empirical studies, assumptions about the likely dynamic productivity gains in several economic sectors.[608] These productivity gains have then been used as inputs into the calculation of the macroeconomic impacts of the AUSFTA on Australia. As can be seen below, the importance of dynamic productivity gains in terms of overall gains in real GDP is significant:

Figure 1- Changes in Real GDP[609]

8.23 The potential benefits arising from dynamic productivity gains have underpinned much of DFAT's process of selling the AUSFTA. The following extract from their submission gives an indication of both the tone, and the significant extent to which the outcomes of the AUSFTA rely on the forecast (direct and dynamic) benefits from investment liberalisation:

According to the CIE's modelling, Australia's annual GDP could be up by around $6 billion (about 0.7 per cent of GDP) as a result of the AUSFTA a decade after the Agreement's entry into force. Total GDP increase over 20 years is expected to amount to almost $60 billion in today's dollars.

Much of this growth will be generated by the dynamic gains expected from the deeper links the Agreement establishes between Australia and the US, with the CIE finding investment liberalisation the biggest contributor to the projected increase in Australia's GDP. But even if these benefits and other 'dynamic' effects of trade liberalisation are excluded, liberalisation of trade in goods and services alone would contribute about $1 billion to real GDP.[610]

8.24 The CIE's assessment of dynamic productivity gains under the AUSFTA has been criticised. In her report for the Committee, Dr. Philippa Dee did not reject the consideration of dynamic impacts altogether, but argued that a more conservative approach would have discussed dynamic impacts without including them in the modelling:

The DFAT/CIE study draws on empirical work that shows that, in addition to having so-called static effects on allocative efficiency, tariff cuts can also have a so-called dynamic effect on sectoral productivity. The study quantifies these dynamics effects by assuming them to be proportional to the size of their tariff cuts.

The studies that the DFAT/CIE study draws on examine productivity levels in Australian manufacturing during a period of substantial unilateral tariff cuts. AUSFTA does not cut tariffs unilaterally, but preferentially- this means that the reductions in price on any given import can be substantially less than the size of the preferential tariff cut.-

The existence of such 'cold shower' effects of tariff cuts on productivity has been hotly debated. Conservative evaluations might note their possible existence, but do not include them in the quantitative analysis.[611]

8.25 Dr Peter Brain of the National Institute of Economic and Industry Research made a similar point:

In terms of a general methodological issue or how the economic assessment should be made, the idea of bottom line point estimates is absurd. I think we can all agree that whatever the dynamic of flow-on effects will be, they will be large-whether it be a multiplier of two, four, six or whatever-compared with the direct effects. Therefore, the analysis should simply focus on the direct effects. This should be done with some humility because there is a wide range of possible outcomes. To accommodate this as best we can, one should try to take into account all possible outcomes in a framework of decision making which allows an assessment of the probable range of outcomes, which we have tried to do.[612]

8.26 Other evidence suggested that, even accepting the inclusion of dynamic effects in the CIE model, the outcomes forecast are far too optimistic. Professor Ross Garnaut, for instance, stated:

The third element of gain is so-called dynamic effects which, if you read the logic of the report, depend on the trade liberalisation being genuinely liberalising. But if the trade liberalisation is not liberalising, if it moves in another direction, then there is no reason to think the dynamic gains will be positive. In fact it is very likely that if the trade liberalisation effects are negative the dynamic defects will be negative as well. So just on a straightforward application of the logic of the models-not the words, and above all not the words of the executive summary, but the logic of the models-suggests that the median estimate of gains would be approximately zero. It would be possibly slightly negative but approximately zero, way below the bottom end of the range of outcomes that CIE and DFAT suggested and put the 95 per cent probability bounds around.[613]

8.27 In other appearances before Senate Committees, officers of the Treasury have been reticent about relying on dynamic effects as a basis for policy decisions, instead adopting an approach more in line with that suggested above by Dr. Brain and Dr. Dee. During an appearance before the Senate Economics Legislation Committee's inquiry into the International Tax Agreements Amendment Bill 2003, Mr Greg Smith, Executive Director, Revenue Group, Department of the Treasury, made the following statement in relation to a double taxation treaty:

The Treasury, perhaps conservatively, but along with longstanding practice and what is also a common practice in many countries-I will not say every country, but in most countries-take the first-round effect and publish that. We draw attention to the other effects-second-round effects, assumption driven effects or other things which we are less certain about-but we have not incorporated them in the official costing- We do a similar thing really, when you think of it, with most of these sorts of things. For example, we may well publish a costing for the research and development tax concession, but of course the purpose of the research and development tax concession is to create dynamic benefits in the Australian economy. We do not publish, and we do not seek to estimate, in our costing what those benefits are.[614]

8.28 Treasury made similar statements directly in relation to the FTA at an Estimates Committee hearing in October 2003:

The long term, significant benefits from this agreement will come from the dynamic interrelationship with the intangibles: competition policy; certainty; work programs on financial services, which will be done through a financial services committee that is being established between the two countries; and work programs on recognising professional qualifications, which is one of the big impediments to service trade at the moment. All of those things are very powerful but they are impossible to quantify, so what you are left with when quantifying the model are things that are easy to measure but perhaps not the most important part of the agreement.[615]

8.29 DFAT, however, rejects criticism of the CIE's assessments of dynamic productivity gains. In a rebuttal of Dr. Dee's comments on this issue, DFAT stated:

Dr Dee rejects the idea of including 'dynamic gains' which flow from the greater competition under trade liberalisation because 'their existence has been hotly debated, and conservative evaluations omit them.' But there is a wealth of econometric evidence which supports the existence of these effects. The CIE has been conservative in its assumptions and has adjusted the magnitude of the gains to reflect the fact that it is a bilateral agreement which has been negotiated.[616]

8.30 The Committee is interested in DFAT's claim of 'econometric' evidence. If such evidence exists and is sufficiently reliable to support this claim, why did the CIE rely instead on assumptions based on empirical evidence?

8.31 It is clear to the Committee that the CIE's estimates of the dynamic benefits of the AUSFTA should, at the very least, be treated with a great deal of caution and scepticism. Instead, the Committee should follow the approach recommended by Dr Dee and Dr Brain, and indeed by Treasury in previous inquiries. The Committee recognises that dynamic effects may result, and may have substantial benefits. However, as Professor Garnaut points out, they may not. Policy decisions in relation to the FTA should therefore be made principally on the basis of the direct effects, with the recognition that dynamic effects may eventuate.

Changes to Foreign Investment Review Board Thresholds

8.32 The Foreign Investment Review Board (FIRB) is a statutory body whose principal function is "to examine proposals by foreign interests for acquisitions and new investment projects in Australia and, against the background of the Government’s foreign investment policy, to make recommendations to the Treasurer on those proposals."[617]

8.33 Currently, FIRB reviews all proposals for foreign investors obtaining substantial interests in Australian companies valued in excess of $50 million, and all proposals by foreign interests to establish new businesses in Australia valued at $10 million or more.[618] If FIRB considers that such a proposal is not in Australia's national interests, it may recommend that the Treasurer block the proposal.

8.34 As such, FIRB is clearly a non-conforming measure under AUSFTA. For it to continue in operation, appropriate provisions must be included in Annex I or Annex II of the AUSFTA. Annex I includes provisions which enable FIRB to continue to operate, but it lifts the thresholds. Under the AUSFTA, FIRB will only review the acquisition of a substantial interest in companies valued in excess of $800 million.

8.35 The Department of the Treasury has indicated that the new FIRB thresholds did not emerge as a result of any perceived problems with FIRB from an Australian perspective. Rather, the new thresholds represented a compromise between the United States' wish to eliminate screening and Australia's wish to retain it:

Ultimately, of course, this was a negotiation, so there was a give and take, if you like. The US were, and remain, very strongly opposed to screening in any form, and they see our arrangements as a potential restriction on investment flows. From our point of view, we felt comfortable raising-and the government’s judgment was that it was comfortable raising-it that far because it achieved the balance between, if you like, the level at which national interest concerns are likely to be real and tangible and, below that level, the level at which they are not likely to be sufficient to warrant the compliance costs associated with the screening.[619]

8.36 The Government has also argued that the raised threshold is essentially a benign change for two reasons. First, while the reduced number of cases referred to FIRB would be substantial, "in terms of value of investment-that is, not the number of transactions but the value of investment-under the proposed arrangements with the US we will continue to screen over 70 per cent of investment by value."[620]

8.37 Second, Treasury pointed out that that FIRB has not been a barrier to US investment in the past. Mr Chris Legg, from Treasury, stated that "I do not think there have been many, if any, cases where US investors have been rejected. There may have been one or two."[621]

8.38 A number of submissions remained concerned about both of these issues. On the issue of the reduction in referrals to FIRB, the Australian Council of Trade Unions noted:

The Office of US Trade Representative estimates that 90% of US investment in Australia over the last 10 years would have escaped screening had the new rules applied retrospectively. Using a three-year retrospective time horizon, DFAT estimates in its Regulatory Impact Statement a reduction in screened proposals by 65-70%. Some commentators have claimed that, under the proposed AUSFTA rules, around 86% of companies listed on the Australian Stock Exchange could be acquired by US interests without being screened.[622]

8.39 The Australian Manufacturing Workers Union argued:

While the AMWU acknowledges that the national interest test has rarely been invoked to prevent foreign investment in Australia, the AMWU notes that the changes would mean that almost 99% of Australian manufacturing companies could be acquired under the proposed AUSFTA with no regard for whether such an acquisition is in the best interests of Australia or Australian workers. [623]

8.40 The Federation of Australian Scientific and Technological Societies made a similar point:

The second area of major concern in terms of allowing for the cherry picking is the change in the threshold for FIRB, lifting its report threshold from $50 million $800 million means that all R&D-intensive science and technology SMEs would fall under that $800 million threshold and there will be no examination. As you are aware, because of the ratchet mechanism, if a future government decided it was concerned about an emerging trend of US firms purchasing Australian R&D companies and taking them offshore, it would not have the capacity to use FIRB as the instrument to address that policy problem.[624]

8.41 In relation to the argument that FIRB has not rejected incoming investments from the United States, the ACTU made the point that:

We acknowledge that the vast majority of applications that are screened by the Foreign Investment Review Board are approved. However, the Board also has a track record of approving applications subject to conditions set to safeguard the national interest. This aspect of the screening process should be borne in mind when it is argued that the current process is simply a time-consuming one that leads to approval anyway.[625]

8.42 It is useful to also note the United States perspective on this issue, which confirms that the increase in FIRB thresholds is likely to increase US investment in Australia, but that the size of this increase will be muted by the relatively open current arrangements:

US industry representatives would also have preferred to discontinue the investment screening performed by Australia's FIRB. However, the minimum size of most foreign investments that require screening has been substantially raised. In general, U.S. investors in Australia must notify the Australian Government (through the FIRB) of investments only if an investment is value at more than A$800 million (US $443.2 million). The previous investment threshold was A$50 million (US $27.7 million). Industry representatives have stated that the higher limits are an improvement in the investment approval process- Industry representatives indicate that due to Australia's fairly liberal existing investment regime, they have been free to invest in most industries despite FIRB screening, and that is not expected to change.[626]

8.43 The Committee notes that there is substantial evidence that the increased FIRB thresholds will have little impact on US investors because the FIRB review process is not currently a substantial barrier to US investment. However, despite this widespread evidence that the impact is likely to be mild, the changes in FIRB arrangements are considered by the CIE as a major factor underpinning the decrease in Australia's equity risk premium (discussed above) and therefore as a major factor underpinning the predicted benefits of the AUSFTA as a whole.

8.44 The CIE report argues that "Australia has relatively high [foreign direct investment] restrictions compared with other countries in the OECD, surpassed only by Iceland, Canada, Turkey and Mexico. These restrictions range from limits on foreign ownership in certain sectors to modest screening procedures for a wide range of investment proposals."[627]

8.45 Under the AUSFTA Australia continues to maintain specific limits on foreign investment in a range of sensitive areas such as newspapers, broadcasting, Telstra, Qantas, airports, and urban land. That is, Australia's 'relatively high foreign direct investment restrictions' on sensitive sectors will not be reduced under the AUSFTA. This leaves only the reductions in the 'modest screening procedures' as the source of increased liberalisation of foreign direct investment- and the impediment raised by this screening process has been shown to be minimal. The AUSFTA will therefore do little to reduce those core factors which result in Australia having 'relatively high foreign direct investment restrictions'

8.46 It is difficult to reconcile such a view with the view expressed above by the United States International Trade Commission that Australia's investment environment is 'already substantially open'.

8.47 Despite all of this, and without any further explanation, the CIE argues that "we are still left with the fact that Australia’s FDI rules are at the ‘restrictive end’ of the OECD scale"[628] and uses this as a justification for assuming these restrictions are responsible for fully half of Australia's current equity risk premium. This provides the basis for a further series of calculations resulting in the claims of a 5 basis point reduction in the equity risk premium under the AUSFTA.

8.48 The Committee observes that the assumption that investment rules account for half of Australia's equity risk premium is at best unproven, and may be substantially exaggerated. Even if the assumption is accurate, the AUSFTA appears (on the above analysis) unlikely to reduce investment restrictions substantially. In either case, there would be flow-on consequences which may reduce the predicted benefits from the AUSFTA by billions of dollars.

8.49 Professor Ross Garnaut criticised the CIE analysis of the impact of the changes to FIRB in these terms, arguing that they failed what he described as the "laugh test":

The Department of Foreign Affairs and Trade has now released the results of the consulting firm’s assessment of the agreement as negotiated. While the main sources of gain in the original estimate resulting from access to highly protected US agricultural markets have disappeared or shrunk dramatically, somehow the total benefits have greatly increased to $5.6 billion. That somehow turns out to be mainly through what are described as back of the envelope calculations of gains, hitherto overlooked from easing FIRB restrictions. There is an air of unreality about this revised estimate. The magnitude of the contribution now attributed to changes in FIRB rules and the basis used to assess it heighten the need for an independent analysis conducted at arm’s length from those whose job it is to sell the agreement to the Australian community.[629]

8.50 In her report for the Committee, Dr. Philippa Dee also took issue with the CIE findings. She argued as follows:

8.51 The Department of Foreign Affairs and Trade continues to support the CIE's views:

In summary, the CIE found substantial gains to Australia from the liberalisation of FIRB restrictions under AUSFTA. These findings are rejected by Dr Dee on the grounds that FIRB liberalisation will not reduce the risk premium on investment in Australia. But there is evidence from investors that complying with its provisions imposes some costs. This reflects the fact that the existence of a screening mechanism may reasonably be expected to have some impact on the level of uncertainty for prospective investors and to contribute to their general perception of risk, notwithstanding genuine efforts to implement the policy in a way that minimises such effects. In addition, other provisions of AUSFTA will improve the investment climate and create added certainty for investors. The impact on the equity risk premium is inherently difficult to quantify: however the sensitivity analysis does indicate that the gains are significant even with a much smaller reduction than assumed in the base case.[630]

Select Committee's view

8.52 The Select Committee, informed the weight of evidence presented during the inquiry, is sceptical of the forecasts made by the CIE and promulgated by DFAT during public debate on the Free Trade Agreement. The loudly proclaimed benefits to Australia arising from a liberalised foreign investment regime and from dynamic productivity gains are based on a series of inferences and educated guesses. The fact that the US International Trade Commission states that the AUSFTA is 'not expected to generate significant new investment' cannot be easily dismissed.

8.53 However accurate the CIE and DFAT predictions may turn out to be, educated guesses do not provide an adequate basis for informed policymaking by parliamentarians, investors or citizens. The CIE report's highly-contested views on investment and dynamic gains have inevitably led to a debate played out in academia, in the media, and before this Committee. The result is that the modelling work which was intended by the government to clarify the impact of the AUSFTA and justify its adoption has in fact led to increased confusion.

8.54 Notwithstanding this, the Committee recognises that the AUSFTA, by liberalising investment between Australia and the United States, is likely to provide a net benefit to Australia in the investment arena. The Committee is far from convinced that the benefit will be as great as that claimed by the Government and the CIE, but it remains likely that a benefit will eventuate. Consequently the Committee does not oppose Chapter 11 of the agreement, but its concurrence occurs despite, rather than because of, the analysis undertaken by the CIE and endorsed by DFAT.