•  About Me 
  •  FAQs 
  •  Mail Me 
  •  Links 
  •  Archives 
February 18, 2011

Submission to the HST Panel

All observers agree that the HST shifted costs on to BC families; how much is in dispute as is whether and to what extent there are any benefits from the shift.

As is the case with much public policy, each option has winners and losers. An assessment of the options needs to identify members of both groups, as well as to attempt to quantify the respective gains and loses. That task is difficult given the limitations of available data prior to the vote. In the absence of hard evidence, no one should confuse opinion, however well intentioned, with fact.

Your panel's terms of reference include four points:

Consumer Impacts

We all have anecdotal evidence of the impact of the HST; the price of a cup of coffee and the annoying addition on restaurant bills are examples. What defies agreement is the average impact for different types of families with different income levels.

Consider what appears to be a fundamental inconsistency in the information provided by the Ministry of Finance. An HST topic box in the September 2009 budget update (pp 77-86), included a section titled "impact on consumers". It said:

Harmonization means three major changes for consumers. Consumers will:
We can probably all agree that the first point is good and the next two aren't so much. The Ministry's topic box went on to estimate the dollar impact of the HST on different benchmark families. Before credits, it said the increase due to the HST for a single individual with a $25,000 income would be $169 per year, and for a single individual with a $80,000 income it would be $427 per year. Other family types with other incomes were shown with impacts between those amounts.

As half of a retired family of two I can say that the impact of the HST on just our daily out-for-coffee habit is over a third of the high end estimate of $427 provided by the Ministry of Finance, and that is before counting anything but our coffee, let alone major costs such as re-roofing.

If the government raises as much from the HST net of rebates as it did from the PST, then the amount saved by businesses must equal the additional amount paid by others. I purchased the Input-Output data from Statistics Canada that the Ministry used to estimate that business will save approximately $1.9 billion per year. Those data substantiate the Ministry's estimate of business savings. Dividing that cost shift by the province's population of 4.5 million yields an estimated average impact on consumers of $422 per person, just $5 less than the ministry estimated for the $80,000 single, but the $422 is an average that includes everyone from infants to seniors.

One way to reduce the estimate provided by simple division so as to come closer to the Ministry's figures is to allow for business cost savings to be passed through to consumers. The topic box said: "It is important to note that in moving to the HST consumers will no longer pay about $2 billion in PST passed on to them by business in the price of goods they buy." If that assertion were true, there would be no cost shift and the Ministry's estimates of impact on various family types should have been zero. The roughly two billion saved by businesses is distributed as follows: $880 million for the construction industry, $140 million for manufacturing, $210 million for the transportation industry, $140 million for the forestry sector, and $80 million for mining and oil and gas. The extent that any savings would be passed through to consumers depends on the properties of each market, but there is no reason to expect BC consumers to benefit from price changes in any of the export markets; consequently, the Ministry's claim about $2 billion in savings from passed through PST is not credible.

SFU's Jonathan R. Kesselman published a paper in February 2011 for the Business Council of British Columbia in which he argues in favour of the HST. He noted that BC businesses had $2.5 billion of PST costs and that $1.9 billion of that was removed (shifted) by the HST. He went on to observe that, based on Statistics Canada's Input-Output tables, "only 52 percent of the embedded taxes are related to BC consumers' purchases, with 33 percent to buyers outside the province, and the remaining 15 percent to public sector bodies (municipalities, universities, schools, colleges, hospitals) and nonprofit and charitable organizations." He added that since the public sector bodies were protected from HST cost shifts by various rebates: "that eliminates $300 million from possible shifting to consumers. As a result, the implied maximum shift of tax burden onto BC consumers would be $1.6 billion ($1.9 billion minus $300 million) if pass-through were zero." He continued saying: "If businesses were to pass through all their tax savings on inputs used to produce items destined for BC consumers, the maximum tax shift to consumers would be only the $600 million share of the removed embedded taxes now borne by exports from the province plus the revenue needed to finance BC's companion changes."

I agree with Kesselman that the range of the HST cost shift is between $600 million and $1.6 billion, plus any costs associated with the change. I disagree with his argument that businesses have quickly passed through the majority of $1 billion in PST cost savings.

Kesselman used consumer price index data through December 2010 for BC and other western provinces to argue inflation was less in BC and that is evidence of nearly complete cost pass-through.

It is arguable whether the CPI is sufficiently robust at the provincial level over a six month period to be used for his analysis. The CPI is made up of over 600 prices, some of which are sampled on only a quarterly basis, for example haircuts and dry-cleaning which are now subject to the HST. Prices are collected over the first three weeks of a month. Unlike many data published by Statistics Canada, the CPI is not revised once it is published. Statistics Canada reports that: "In general accuracy is better at the Canada level for any product index in the CPI compared to the same index at the province or city level. Also, accuracy is better at the All-items or major component levels of the CPI compared to individual product indexes. Finally, the CPI is more accurate as an indicator of change over several months or a year compared to the accuracy of the price change measured from any one month to the next."

The day after Kesselman's article was made available on the Internet, the Vancouver Sun published an opinion column from him which promoted his conclusions. Unfortunately for him, that same morning Statistics Canada made available the January CPI figures. By adding just one more month, the ratio that is key to Kesselman's work was reversed. He based his argument on the all items excluding energy index decreasing by 0.089 per cent between July and December, but when January is added the numbers show it increased by 0.089 per cent between July and January.

In addition to a table showing changes in various components of the CPI for BC, Kesselman prepared a table adjusting the BC CPI changes relative to a population weighted average of the corresponding indices for Alberta, Saskatchewan and Manitoba. He suggested the result adjusts for underlying inflation so as to isolate the change in BC due to any price reduction flowing from HST input-tax-credits. Of course, some businesses only file HST returns on a quarterly or annual basis so Kesselman must have presumed that they accurately anticipate their returns. More importantly, the difference between percentage CPI changes in BC and the other western provinces could be explained by many things other than what Kesselman attributes to them. A proper model would have to adjust for all other factors, but Kesselman's approach was to do a simple calculation with no underlying model. Applying Kesselman's method to the CPI data since 1981, shows that, for six month periods of price changes, BC has had slightly higher inflation than the weighted average of other western provinces for most months since July of 1996. The point of that comparison is simply to show that Kesselman's calculation cannot be interpreted as he chose to do. Kesselman tried to read far too much into data that aren't sufficiently robust to bear the weight of his argument.

Notwithstanding issues with the CPI, and with no test of significance, on the basis of comparisons on tenths of a percentage point change in the provincial CPI, Kesselman would have us believe that almost $1 billion in businesses cost savings due to the HST have been rapidly passed through to consumers in the form of many small price changes. Kesselman wrote: "Public skepticism on this point is widespread. Admittedly it is difficult for consumers to identify small price cuts across a vast array of goods and services-whereas it is easy to see the HST, which is printed clearly on every sales receipt."

Compare Kesselman's analysis to one of the first studies posted to the government website in support of the HST, from the 2007 C.D. Howe Institute, Commentary, No. 253 by Michael Smart who analyzed the implementation of the HST in the Atlantic provinces. Smart reported (p. 12) that: "CPI prices fell by about 0.3 percent in HST provinces after 1997, compared to the corresponding change in RST provinces. This difference is statistically insignificant but extremely close to the estimated 0.5 percent reduction in taxes under the reform." In this context "statistically insignificant" means that there is an equal chance that the implementation of HST could have caused prices to go up or had no effect at all. In other words, nothing was proven. The study went on to report details on 8 components of expenditures, with a footnote regarding statistical significance. Five of the eight components of expenditure were not significant at the 5% or less level. Shelter showed a statistically significant (1% level) increase of 1.4%. That is one of the reasons the study concluded that the implementation of HST was "mildly" regressive. HST applied to fuel in those provinces which might account for that result.

Unfortunately, evidence on whether HST savings for businesses are passed through to consumers or not appears to be a topic that should be regulated to the journals of insignificant and unworthy to report results; but that doesn't stop advocacy groups from quoting and using those results.

Impact of each option to B.C. businesses and B.C.'s economic competitiveness

An increase in relative prices for most of the service sector and for previously PST exempt goods sectors is not helpful for those businesses. The extent of damage to those sectors is an empirical question which is challenged by lack of data. Statistics Canada reports monthly on food service sales, but with a lag, revisions for the previous three months and annual revisions. Apart from the next round of annual revisions, the latest "final" monthly data are for August 2010, giving us virtually no reliable data for measuring the impact of the HST on restaurants. Preliminary data for November 2010 show a 2.8% increase relative to November 2009 for BC compared to a 5.2% increase for all of Canada, but those figures are subject to several rounds of revisions. Data for various components of the service sector are harder to find and less reliable. It appears that British Columbians will vote on the HST with no objective evidence of its impact on the sectors where it increased relative prices

Like the service and previously PST-exempt goods sectors that suffered some harm from the HST, there is no hard evidence on the sectors that may have benefited, although there is an abundance of rhetoric. The government's HST website claims: "The HST saves businesses money by reducing operating, investment and administrative costs. Without the HST, other jurisdictions with a VAT will have a competitive advantage over B.C." Ironically it went on to comment on our closest neighbour: "The HST is a Value-Added Tax. VAT exists in every OECD (Organisation for Economic Co-operation and Development) member country except the United States." The government website should have added that Alberta, Saskatchewan and Manitoba also do not have a harmonized sales tax.

Whether a sector achieves a net benefit from the input-tax credits that are part of the HST depends on how those credits compare to the effect of discouraging customers as a result of higher prices. That comparison varies depending on the capital intensity of each industry and its price-elasticity-of-demand. Many service oriented businesses have few inputs for which they can claim an input-tax-credit, yet they lose more than one percent in sales for every one percent increase in price. Of course there are thousands of cases for various businesses, and for most of them there is little or no publicly available data for comparing input-tax credit savings with lost sales. For the entire BC restaurant sector, Statistics Canada's input-output table revealed potential input-tax credit savings of $48 million in PST for 2006 (less than one half of one percent of total industry revenue). That means restaurants get very little by way of input-tax credits, but their customers pay 7% more. Other industries that were previously exempt from the PST on most of their inputs are in a similar position, net losers with respect to the HST.

It is clear that export-oriented, largely capital intensive, industries benefit from the HST, by $600 million per year according to the figures provided by Kesselman. Is that tax break sufficient to make a difference in investment in those industries? In 2002 the BC government eliminated the PST on "production machinery and equipment"; however, the Ministry of Finance has not been able to produce any studies showing increased investment as a result of that tax change which the July 2002 budget update documents estimated to cost $134 million per year. Apparently the argument is that lowering costs in general will stimulate investment.

Data on investment by province come from the provincial accounts published by Statistics Canada on an annual basis; more frequent data are not available. Gross business capital formation consists of residential structures, non-residential structures and machinery and equipment. Investment in machinery and equipment is considered the most important as it can embody technological change.

In his 2007 study, Smart attempted to argue that the change to the HST encouraged investment but his data included offshore oil and gas development. A comparison of gross business capital formation in machinery and equipment for the provinces since 1981 shows that over the long term investment in BC grew at the same pace as investment in New Brunswick and Nova Scotia. In the past ten years, investment grew faster in Newfoundland and Labrador than in BC due to the offshore.

The BC Progress Board has identified business investment as one of its important benchmark performance indicators. The Board's 2010 report observed that among the provinces in 2009 BC ranked fourth in terms of total business gross fixed capital formation as a percentage of GDP, and sixth in terms of non-residential gross fixed capital formation. Its last report noted: "British Columbia is a middling to strong performer in Total Gross Fixed Capital Formation (GFCF) as a percent of GDP when compared to OECD countries. BC's ratio of 23.4 percent in 2009 earned it a fifth-place rank, down one place from its position in 2008. Canada ranked eighth of 31 with a ratio of 21.5 percent." That comparison is particularly important since the OECD countries have value added taxes. If that gives them an advantage in attracting investment, it is not obvious from the data presented by the Progress Board.

Claims about the HST creating investment and job growth are nothing more than claims.

All that can be said for certain is that the HST gives a cost break to some businesses with the break being a greater proportion of cost the lower labour is as a proportion of total costs, since labour is not subject to the tax or its credits; the elimination of the HST and return to the PST would give most service and previously PST exempt industries a 7% price advantage relative to the HST system.

Fiscal implications of each option to the provincial budget in both the short and long term

The government has emphasized that the HST is revenue neutral from its perspective. That means it expects to raise the same amount from both taxes. With the HST enough is raised so that after rebates and associated tax changes, the net amount is roughly the same as what was raised by the PST. I see no reason to quarrel with the government's assertion; however, it is important to understand that in making that statement the government is including a package of tax cuts and rebates that need to be fully listed in order to understand exactly what makes the HST revenue neutral to government.
The Ministry of Finance has experience with changes in the realm of hundreds of millions as part of the federal-provincial income tax agreement. BC can also expect large retroactive adjustments to its HST revenues. The Ministry of Finance should be able to advise whether those fluctuations are expected to be worse than what it experienced when administering the PST.

Some economists suggest that in the long run HST revenues would grow more rapidly than PST revenues because of the broader tax base, but that would take a very long time. For example, if the PST base grew at the annual rate of 5.5%, after 10 years it would increase by 71%. If 83% of the HST base were the same as the PST base growing at 5.5% and 17% were previously excluded goods and services growing at 6.5%, after 10 years the HST base would increase 74%. That difference does not appear to be sufficient to justify a tax shift of between $0.6 billion and $1.6 billion per year forever.

One of the impacts of reverting to the PST will be the cost of restoring a branch capable of administering the tax. The 2011-12 budget documents tabled in the legislature on February 15th report that: "In 2011/12, the forecast assumes $30 million in PST revenue representing expected audit collections in respect of prior years." Apparently all of the former PST administration has not been dismantled; nevertheless, reinstating a department capable of administering the tax will not be trivial.

Of course, the transition payment of $1.6 billion provided by the federal government will have to be repaid if the HST is killed prior to the exit provisions in the federal-provincial agreement. The terms and timing of any repayment would be subject to negotiations, but in the worst case scenario it should be thought of as accurately re-stating what the provincial debt would have been had British Columbians not been surprised with the HST announcement just two months after the last election.

Relevant information and analyses from other jurisdictions

In July 2009, the same month Premier Campbell announced BC's HST, France reduced its value added tax (VAT - the equivalent of our GST or HST) on restaurant meals from 18.5% to 5.5%. It did that in the hope of creating an estimated 40,000 more restaurant jobs.
Proponents of the HST are fond of referring to the number of countries that use the tax and to an untold number of economists who are alleged to support it. Since the HST can probably be credited with destroying the Campbell government, and since all candidates for leadership of the BC Liberals declared support for the HST, it is reasonable to presume that the government's HST website would provide the most persuasive literature available in support of the tax. Look at the list of 26 articles under "reports and studies" on the government website. Not one is from a top-level peer-reviewed journal; what's offered is from "think tanks" and advocacy groups. They are entitled to their opinion, but it should be recognized as opinion, some deserving publication in a journal of insignificant results.


The HST shifts taxes from businesses to consumers. Estimates of the size of that shift are a minimum of $600 million per year and as much as $1.6 billion per year. The precise impact on each family varies by consumption patterns and income.

Claims with respect to improved investment and employment as a result of the HST are difficult to substantiate. France cut its value added tax so as to stimulate its restaurant industry.

Whatever costs and benefits are attributable to the HST, those that benefit are not the same as those who bear its costs. The tax shifts costs and benefits.

Re-instating the PST by returning to the tax system that existed on June 30, 2010 will not be easy but relative to a permanent annual cost shift of a minimum of $600 million and as much as $1.6 billion, many will probably consider the transition cost worth the change.