Tag: The Treasury

Warren Tease, Principal Adviser, Macroeconomic Conditions Division spoke at CEDA’s Economic and Political Overview Conference about forecasting in Treasury.

Most forecasts for growth in most economies have been persistently too high over the past few years.

In that time, a key feature of the forecasts for the global economy and for Australia has been the persistent downgrading of forecasts. The pattern of downgrades is clear. In almost every year since the crisis, forecasters have been too optimistic about the outlook for growth. Treasury’s forecasts have exhibited the same behaviour. This persistent bias is perhaps the most unusual attribute of recent experience, at least in Treasury’s case, as past forecasting reviews have shown that Treasury’s real GDP forecasts have not exhibited persistent errors over time.

Digging a little deeper into this quantitative work I would also observe that Treasury’s forecasts did well in normal times but were challenged when conditions were unusual.

Thus, while there was no evidence of persistent errors over a very long time period, Treasury’s forecasts were punctuated by persistent and sometimes large errors over horizons relevant for policy makers and their advisers.

This can be seen in the following chart which compares Treasury’s forecasts of nominal GDP growth with actual outcomes since the early 1990s. I use nominal rather than real GDP as it is a key input into forecasting budget outcomes.

Chart 2 — Nominal GDP Forecasts and Outcomes

Source: Budget papers and ABS cat. no. 5206.0

It shows three distinct periods of forecast errors. Forecasts regularly overestimated nominal GDP growth in the early 1990s and after the financial crisis while understating outcomes for most of the 2000s prior to the crisis.

Each of these episodes coincided with Treasury persistently erring in its inflation (measured by the GDP deflator) forecasts. This observation has been made in each of the previous forecast reviews.

As previous reviews noted Treasury missed the decline in inflation in the early 1990s.

More recently, the forecasts did not incorporate fully the impact of the unprecedented rise then fall of commodity prices. In the upswing, Treasury’s forecasts did include projections that prices would eventually fall but these proved to be wrong as prices continued to increase. In the downswing, prices have fallen more rapidly than Treasury’s assumptions implied. Heightened volatility in Australia’s terms of trade has made it more difficult to forecast nominal GDP outcomes.

While difficulty in forecasting the GDP deflator has been a feature of Treasury’s (and others’) forecasts, past reviews have shown that its real GDP forecasts were more accurate and unbiased. Therefore, it is the persistent errors in Treasury’s real GDP forecasts of the past few years that have been unusual.

The chart shows that in the 1990s and 2000s there was no distinct under or overestimate in Treasury’s forecasts for the real economy. However, since the crisis Treasury, like the other forecasters presented earlier, has for the most part overestimated real GDP growth.

Chart 3 Real GDP Forecasts and Outcomes

Source: Budget papers and ABS cat. no. 5206.0

Now I do not wish to push this point too hard as the sample set is very small but it is consistent with what we are seeing across a range of forecasters and countries.

He has reviewed the forecasting approach, and concluded that in Australia a smaller number of input metrics have been used, compared with other countries. In addition Australia, as an open economy is wide open to international factors which impact locally.

“For a small open economy like Australia, persistent and short-term deviations from trend are likely to be driven by global economic shocks, large changes in commodity prices or from the financial sector. Other factors could also be important but these three factors encompass much of Australia’s experience with shocks.”

Finally, there is a bias towards trend performance, so any deviation is harder to cope with.

“It means that Treasury’s forecasting approach is likely to generate reliable forecasts at times when economic conditions are normal but will be challenged at other times.

There is a therefore a high probability that structural changes or persistent shocks will not be adequately incorporated into the forecasts”.

Good to see a discussion about the complexity of forecasting in a volatile environment. We should not be taking their projections as gospel truth. Forecasting ain’t easy!

A significant speech by Nigel Ray, Deputy Secretary, Macroeconomic Group, The Treasury.  The latest data on the population and labour force trends suggest that potential output will be lower than estimated at Budget. So potential GDP will grow by around 2¾ per cent over the next few years, lower than the 3 per cent estimated at Budget.

There was criticism of the Budget that the projections for real GDP growth over the medium term were too high.

Some have suggested these projections were raised in order to improve the Budget position. Can I say at the outset that this claim is simply not true. In fact, when Treasury introduced the current methodology in the 2014-15 Budget, we published analysis in a separate Working Paper showing that the impact of these changes on the underlying cash balance was very small – an increase of just one-tenth of a per cent of GDP by the end of the medium term.

It is true that Treasury’s framework assumes that the economy will grow faster than potential for a period into the medium term.

This reflects our estimates that the economy is currently operating well inside the limits of its productive capacity, following a period of below-potential growth over recent years. This view is supported by the slow growth in wages and prices that we are currently seeing.

We expect the economy will still be operating with excess capacity by the end of the forecast period in June 2017, even factoring in a modest pick-up in growth over the next two years.

In these circumstances, we assume that necessary adjustments in prices and wages will see the economy’s spare capacity absorbed and the unemployment rate decline towards the non-accelerating inflation rate of unemployment (NAIRU) over a number of years. Similar approaches are used by the Congressional Budget Office in the United States and the UK’s Office for Budget Responsibility.

Our previous approach was to assume that the unemployment rate returned to the NAIRU in the first year of the projection period. This assumption had the considerable benefit of simplicity. And so it had merit when the unemployment rate was not far from the NAIRU. But it was (rightly) criticised as being an unrealistic assumption in circumstances where the unemployment rate was expected to be significantly above the NAIRU by the end of the forecast period.

And in such circumstances, the methodology suffered from internal inconsistencies. While the unemployment gap was assumed to close, the output gap was implicitly retained.

Another alternative is to assume that an output gap will exist indefinitely, and that the unemployment rate will remain above the NAIRU. Apart from also being unrealistic, the assumption that cyclical unemployment will last indefinitely would seem odd given that policy settings are designed to avoid this very thing.

Instead, our approach is to assume that the economy’s output gap will be absorbed over a period. History suggests this takes about five years. For simplicity, we assume that the adjustment takes place evenly over these five years.

By definition, the closing of the output gap requires real GDP to grow faster than potential over this time.

While involving some necessary simplifications, this approach is consistent with historical experience where the economy has tended to grow above potential for a number of years in a row following a period of sustained weakness.

Chart 10: Real GDP growth

 

Source: ABS Cat. No. 5204.0 and 2015-16 Budget forecasts.

To implement the methodology we need estimates of the level of potential GDP now and into the future. Potential GDP is not a quantity that we – or anyone else – can directly observe, either now or over history. Instead, we have to use the best methods at hand and the best available data to estimate it. To do this, we estimate trends and make assumptions about each of the economy’s supply side drivers – population, participation and productivity.

Once we have estimated the current level of potential GDP, the next step is to project how potential GDP will evolve over time – that is, we estimate the future growth rate of potential. Again, we do this for each of the 3Ps separately.

Population projections are based on the latest population figures from the ABS and net overseas migration estimates from the Department of Immigration and Border Protection, together with our estimates of fertility and mortality. Projections of the trend participation rate and average hours are built up from age and gender-specific labour force data.

For labour productivity, we assume trend growth of 1.6 per cent, which is the 30-year historical average and is consistent with our approach in the Intergenerational Report.

But none of these numbers is set in stone.

And like similar national economic advisers overseas, Treasury continually reassesses its estimates of potential growth and the size of the output gap in light of new data and new methods.

New evidence on the size of the population suggests that Australia’s current productive capacity is slightly lower than we estimated at the time of the Budget in May.

Australia has experienced rapid population growth over the past ten years, fuelled by solid increases in net overseas migration. This came on the back of our relatively strong economic performance and low unemployment compared to other countries. But as economic growth has softened, recent immigration outcomes have come in lower than the Department of Immigration and Border Protection initially expected. This mainly reflected declines in temporary visa holders and lower net migration from New Zealand.

Chart 11: Revisions to growth in working-age population

 

Source: ABS Cat. No. 6202.0 and Treasury.

The revised data from the ABS – shown in the blue line in the chart – showed that growth in the working-age population slowed to 1½ per cent over the year to June 2015, well below the Budget assumption of 1¾ per cent and for that matter well below its average yearly growth over the past ten years (also 1¾ per cent). Like the Reserve Bank, we expect Australia’s population to continue to increase at around its current, slower rate of growth over the next few years.

This has immediate and unavoidable consequences for the economy’s potential: fewer people means a smaller supply of employees. So for a given capital stock and level of productivity, we can now produce less output overall than we previously estimated.

We have also reconsidered the contribution of average hours worked to potential GDP going forward In light of new data. The latest quarterly data suggest that a larger portion of the decline in average hours reflects trend rather than cyclical factors, implying that the previous recovery in average hours factored into our projections may be too large. Our revised estimates of trend are shown in the solid red line in the chart.

Chart 12: Average hours worked – trend and cycle

 

Source: ABS Cat. No. 6202.0 and Treasury.

Taken together, the latest data on the population and labour force trends suggest that potential output will be lower than estimated at Budget. We now think potential GDP will grow by around 2¾ per cent over the next few years, lower than the 3 per cent estimated at Budget.

It’s also worth noting that the ageing of the population means that the economy’s potential growth rate is projected to fall slowly over time, reaching 2½ per cent by 2050.

As a consequence of these changes, the output gap will be smaller than estimated at Budget. The chart shows our latest estimates of the history of the output gap, compared with Budget.

Chart 13: Output gap estimates

 

Source: ABS Cat. No. 5206.0 and Treasury.

We will finalise our forecasts for MYEFO after the national accounts are released next week and in light of our latest business liaison round, which we completed last week. Those forecasts will inform our estimate of the output gap at the end of 2016-17 and so the pace at which the economy will need to grow to close it over the following five years.

But it is likely that, on average, we won’t have to grow quite so fast to get there as we previously thought. The economy will of course, by definition, still have to grow a bit faster than potential.

It needs to be remembered that Treasury forecasts the economy predominantly to inform the budget. While our analysis of the economy also informs policy development more broadly, it is the need to prepare estimates of revenue and expenditure over the forward estimates and beyond that drives our approach.

And it is nominal GDP that matters for revenue.

When unemployment is above the NAIRU, it is reasonable to expect that wages and prices will grow more slowly than usual. That effect was one of the main reasons why the change in methodology in 2014 had a smaller fiscal impact than many commentators expected.

The slower price growth drags directly on the dollar-value of the economy’s output, muting the effect on revenues of strong growth in output volumes.

And whereas the economy fully recovers lost ground on output volumes over the assumed period of cyclical adjustment, this is not the case for nominal GDP. The dollar value of the economy’s output that is lost to slow price growth over the medium term is lost forever.

Changes in population, prices and the number of people unemployed will all affect estimates of expenses. The Government will publish revised estimates of the projected medium-term fiscal position in MYEFO. Compared with the Budget base, it is reasonable to conclude that the changes to population and hours worked that I have outlined would, everything else equal, lower the path of the projected underlying cash balance.

It is important to reiterate that this all rests on our economic projections, which are based on a set of assumptions about how the economy works:

• assumptions about the supply-side drivers of Australia’s productive capacity – about population, participation and productivity; and
• assumptions about the time it takes to return the economy to potential output through adjustments in prices and quantities in the labour market.

There are a number of reasons why our projections could be wrong.

The output gap might be smaller than our estimates. Our revised projections reflect the recent downgrades to historical population data. But if trend participation is lower than our estimates, this might mean potential output, and the output gap, is smaller than the projections assume. As a result, the economy might not need to grow as fast to return to potential. Present conditions in the labour market – the wage price index is growing at its slowest pace on record – and measures of underutilisation suggest there is a large amount of spare capacity in the economy. To us, this signals that there is currently a large output gap even when we factor in lower population growth.

The NAIRU might be higher than our estimated 5 per cent. This would also narrow the output gap as it would erode Australia’s available labour resources. Our experience during the mid-2000s provides some reassurance that the NAIRU is about 5 per cent and that we can operate with an unemployment rate around this level without inflation rising too much.

On the other hand, long periods of high unemployment, and in particular, widespread long-term unemployment, could cause a structural lift in the non-accelerating inflation rate of unemployment. You might expect this outcome in periods of excessively high unemployment where the long-term unemployed lose skills and their connection with the labour market. However we are not now in a recession. And Australia’s current long-term unemployment rate is nowhere near what it was in past recessions.

Productivity growth could be slower than it has been in the past. Our projections assume labour productivity will grow in line with its average rate of growth over the past 30 years. Productivity could grow more slowly because of structural shifts in the economy. As services become a bigger part of our economy, productivity growth could ease. Equally, breakthrough innovation and the realisation of productivity gains from current technologies could see productivity grow faster than it has in the past.

This heavy reliance on this set of assumptions highlights the need for Treasury to continue to expose our frameworks and assumptions to outside scrutiny.

We regularly check our methods against the best approaches used by forecasters both domestically and overseas. And our methodologies are published in working papers that are freely available online – this includes our methodology for the medium-term projections. We are releasing today a short note that sets out in more detail our updated estimates of potential growth and the output gap.