ࡱ> Root Entryg;AA F@Cs ?5sAWordDocumentr ;AA`BqObjectPoolw;AB@Cs@CswSSummaryInformation(;-B[B,- 0#C!".@/81234567=9;<I>?BAGDEF NJKLMQOP:^_`abcdeCompObj#fObjInfo"Equation Native  lCompObj2jRoot Entryg;AA F@Cs ?5sAWordDocumentr ;AA`pObjectPoolw;AB@Cs@CswSSummaryInformation(;-B[B-, 0$C!"%&')*+H@1234567=>?BAGDEF NROP(STUVWXYZ[\]ܥhc teIBqpJTTP`P`P`P`P`b`:bbb$b:bb]p1Vb:b"bbbbbbdddd=VdegpXp\gP`b%&bbbbgcP`P`bVbcccb$P`bP`bd ?5sd`,`JP`P`P`P`bdcAcImplications of the Barnett Formula: or The Bomb under the Scottish Parliament Dr J. R. Cuthbert Introduction. 1.) In the devolution White Paper, the Government announced that the primary mechanism for the control of the finances of the Scottish Parl՜.+,0HPt|  Dell Computer Corporation %Implications of the Barnett Formula:Oh+'0  <H p |  and, then equal per capita increases in Scotland and England will result in a lower percentage increase in Scotland.) In fact, the squeeze implied by Barnett is proportional to the ratio of the per capita spends in the two countries. Specifically, if, for a given program, the per capita spends in Scotland and England are PCS and PCE respectively, then the percentage increase, y, implied by an x% increase in England is given by the formula  EMBED Equation.2  . 6.) The theory of the Barnett formula, therefore, implies that overall levels of public expenditure per head in Scotland will converge with those in England, (assuming a rising cash trend in public expenditure). As originally applied, however, the Barnett formula represented a rather weak discipline on overall levels of Scottish public expenditure. There were a number of reasons for this: in particular:- (a) Since the Public Expenditure survey is a rolling programme, a new end year has to be established each year. In the early years of Barnett, this was done by uprating the previous end year by a common percentage uprate for both English and Scottish programmes. This had the effect of sidestepping the Barnett discipline, of a common per capita increase, for a major component of the increase in public expenditure each year. (b) Major public sector pay awards, (like the teachers and health service workers), were funded in full, irrespective of Barnett. (c) The population ratio upon which Scotlands Barnett share was based was not uprated for a number of years. Given that Scotlands population has been declining relative to England, this was in Scotlands favour. 7.) Over recent years, however, the Treasury has been quietly, but effectively, tightening up on the way Barnett operates. In particular:- (a) Since 1993, the new end year in the Public Expenditure survey has been determined by simply taking the previous final year cash total, without any general percentage uprate. This means that all increments for Scotland are now subject to Barnett. (b) Since 1992, the practice of fully validating major pay awards has ceased. (c) The population factor used in applying Barnett was updated in 1993 to reflect 1991 Census information on actual population changes: ( under the current proposals, it is proposed that the factor is updated every year.) The important factor to note is that the Barnett formula now has teeth, in a way it did not have originally, and that it is already delivering, and will continue to deliver, a squeeze on Scottish public expenditure. The Magnitude of the Implied Squeeze on Scottish Public Expenditure. 8.) As implied by the formula at the end of para.5 above, if the Barnett formula were simply applied to comparable public expenditure programmes in Scotland and England, then the size of the squeeze in Scotland is related to the ratio of the per capita expenditures in England and Scotland. Figures of relative public expenditure in England, Scotland and Wales are published in the annual Public Expenditure Statistical Analysis. The figures are not entirely satisfactory, since certain elements of expenditure, which cannot be readily attributed to the constituent countries, are excluded. However, the figures do give a reasonable first approximation to the magnitude of the squeeze on Scottish expenditure implied by Barnett. Excluding Social Security and Agriculture from the figures, (since these are not in the Scottish Block), the 1997 Public Expenditure Statistical Analysis implies that the ratio of per capita identifiable public expenditure in Scotland, compared with England, was 1.29 in 1995/96. Applying the formula in para.5, (and making a small additional adjustment to allow for the fact that the Scottish Block is slightly smaller than Identifiable Public expenditure), this implies that an overall increase of x% in public expenditure on the relevant programmes in England will yield, via Barnett, an increase of .81x% in the Scottish Block. Table 2 shows the figures on identifiable per capita public expenditure broken down by major service: this gives an indicative attribution of the squeeze which would apply if the Barnett formula were applied at individual programme level. 9.) The analysis in the previous paragraph, while interesting, is nevertheless not fully adequate, particularly as regards the implications for the funding of local authority services. One contributory factor to this is that, as already noted, the figures for identifiable public expenditure exclude loan charges paid by local authorities. A more detailed analysis of the funding requirement for local authority current expenditure reveals an even more worrying picture. 10.) It is useful to begin this analysis of local authority current expenditure by distinguishing three different ways of measuring expenditure, as follows:- (a) The starting point is local authority gross service expenditure: i.e., the gross amount which authorities actually spend on the provision of services. (b) However, local authorities generate a certain amount of income from fees and charges. For public expenditure management purposes, this is netted of from gross service expenditure: thus, the public expenditure figures which feed into Barnett are actually net of fees and charges. (c) Finally, local authorities also pay loan charges out of current expenditure: so local authorities total current funding requirement is, apart from minor items, the sum of net service expenditure plus loan charges. The following table shows the per capita figures for each of these three measures of expenditure in Scotland and England. Local Authority current expenditure per head: 1994/95 Scotland England RatioS/E Gross service expenditure 1222 989 1.24 Net service expenditure 1075 814 1.32 Total funding requirement 1231 895 1.38 (Source: derived from Comparative study of local authority current expenditure in Scotland, England and Wales Coopers and Lybrand/PEIDA : Table 2) Note how the ratio in the final column of this table increases. This is because:- (a) English authorities raise proportionately more in fees and charges than Scottish authorities: (hence the difference between the ratios of gross service expenditure and net service expenditure). (b)Scottish authorities have higher loan charges than English. The important implication of the table is that, while Barnett operates on the middle line of the table, the effective squeeze in Scotland as regards local authority current services relates to the total funding requirement, and will therefore be more severe than implied by the relative levels of net service expenditure. This squeeze on local authority current expenditure will work its way out in terms of pressure on council tax rises- as will be seen in the next paragraph. 11.) Suppose that the Government decides that there should be an x% increase in overall spending in England, funded by equal percentage increases in council tax and Government support to local and central programmes. As has been seen, the percentage increase in the Scottish Block will be approximately .81x. Unless the Secretary of State is willing to squeeze other programmes within the block, this means that Government grants to support local authority current expenditure will also rise by .81x%. Now, in Scotland, 12.6% of the total funding requirement for local authority current expenditure is provided by Council Tax, and 87,4% by Government grants. So to achieve an x% in total funding for local authority current expenditure, the required percentage increase, y%, in council tax would need to satisfy (.126)y + (.874)(.81)x = x . In other words,  EMBED Equation.2  This is a startling figure. The implication is that, to achieve the same percentage increase in total funding in Scotland as in England, the percentage increase in council tax in Scotland has to be almost two and a half times that in England: or else other parts of the block will have to be progressively squeezed. 12.) There are a number of important points to be made about the above:- (a) First of all, the above analysis is based on a description of the existiܥhc deIp#JTT`````a`&aaaaaap1a:,b"NbNbNbNbNbNbIdKdKdKd=d"fgHpXp\g=`Nb%&NbNbNbNbgd``Nbaiament would be the continuation of the current Barnett formula. This paper examines the Barnett formula, and some of the resulting implications. The conclusion is that the proposed arrangements pose a severe threat to the succesful operation of the Scottish Parliament. The Public Expenditure Control Total, and the Barnett Formula. 2.) The Governments targets for the size and growth of public expenditure are expressed in terms of the control total. The bulk of the control total represents public expenditdddNb`Nb`NbIdbՋs`,@`J````NbIddAdImplications of the Barnett Formula: or The Bomb under the Scottish Parliament Dr J. R. Cuthbert Introduction. 1.) In the devolution White Paper, the Government announced that the primary mechanism for the control of the finances of the Scottish Parliament would be the continuation of the current Barnett formula. This paper examines the Barnett formula, and some of the resulting implications. The conclusion is that the proposed arrangements pose a severe threat to the succesful operation of the Scottish Parliament. The Public Expenditure Control Total, and the Barnett Formula. 2.) The Governments targets for the size and growth of public expenditure are expressed in terms of the control total. The bulk of the control total represents public expenditure on services: and the public expenditure programs of the Whitehall service departments form sub-components of the overall control total. Not all of the control total, however, is allocated to service departments: for example, the debt interest paid by general government, (including local authorities), is within the control total, but is not allocated to service departments. (as will be seen later, this has important implications for Scotland.) 3.) Scottish departments have a single Scottish component wi%Implications of the Barnett Formula:DCPreferred CustomerA NormalPreferred Customer24Microsoft Word for Windows 95@h3@TXh|s@6{r@HԨs a=(ր(iJmJiJ y = [1-(.874)(.81)].126x    =2.32xEquation Native @XCompObj5fObjInfo4 !$&'()*+,-.03689:;<=?A FMicrosoft Word Document MSWordDocWord.Document.69q FMicrosoft Equation 2.0 DS Equation Equation.29q E  .1   &  & MathType "-@Times New Roman\ - 2 j>x 2 <1.34` 2 8 ` 2 =2 H .75x%``> & "Systemn-L E+l<mJ iJ4nJ x1.34 = .75x%ure on services: and the public expenditure programs of the Whitehall service departments form sub-components of the overall control total. Not all of the control total, however, is allocated to service departments: for example, the debt interest paid by general government, (including local authorities), is within the control total, but is not allocated to service departments. (as will be seen later, this has important implications for Scotland.) 3.) Scottish departments have a single Scottish component within the control total. Apart from a small element relating to the Forestry Commission, this relates entirely to services within the responsibility of the Secretary of State for Scotland. For the great majority of these services, the Secretary of State has discretion as to how the services should be delivered, and can determine expenditure priorities between services. Expenditure on these services is grouped into an aggregate known as the Scottish Block. (The exempted services, which are within the Secretary of States responsibility, but not in the Scottish Block, are services where decisions are still essentially made at GB level- principally agriculture.) Table 1 shows the composition of that element of the control total relating to Scottish departments, and how this breaks down to give the Scottish block. 4.) Each year, as part of the annual Public Expenditure round, the Government determines any necessary changes to the overall control total, and to the public expenditure programmes of the individual Whitehall spending departments. Typically, this involves a series of bilateral discussions between the relevant Minister, and the Chief Secretary to the Treasury. Changes to the Scottish Block, however, are determined simply by the application of an agreed formula: this is the so-called Barnett formula, introduced in 1978 to provide the basis for funding the devolution proposals then current. Under the Barnett formula, the Scottish Block is changed by the pro-rata population share of the change to the relevant English programme: or, putting this another way, the per capita increase to the Scottish Block is the same as the per capita increase to the English programme. 5.) The implication of the Barnett formula is that, if public expenditure levels are rising in money terms, then this will in due course cause convergence in the average levels of public expenditure per head in England and Scotland. (This can be readily seen, since, if the per capita expenditure on a given programme is higher in Scotland than EnglDDDDDDDDDDDE-E[ExEEEEEEF-FJFvFFFFGGGGJGxGyGGGG H9H^HHHHHH~Ip p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p -~IIIIIIIIIIIIIIIII6URUTUZZZZ]]]]]p p p p p p p p p p p p p `%K@Normala "A@"Default Paragraph Font @ Footer 9r )@ Page Number expenditure on services, and other items like general government debt interest. Twhilethe remainder is entirely for which he is responsibleT 2 .A. 4 4s in the UK. The control total includes the public expenditure programmes of the Governments service departments, e.g., Health, Education, etc.qw\#IIIIIIIIITTTTTTTUU4U5U6U7UPUQURUSUTUUUVU]U`UaUbUUUUUUUUUUUUUUXLX[XXXXXYGYHYXYYYZYaYkYlYxYyYYYYYYYYYYZZZZZvZZZZZZZZZZZuD8ceKuD8avK uDcPaucXDocumentSummaryInformation8 _955275415F@Cs æCs_955283892F æCsͦCsOle _955368158F`sqys|sOle BPIC >LMETA 7thin the control total. Apart from a small element relating to the Forestry Commission, this relates entirely to services within the responsibility of the Secretary of State for Scotland. For the great majority of these services, the Secretary of State has discretion as to how the services should be delivered, and can determine expenditure priorities between services. Expenditure on these services is grouped into an aggregate known as the Scottish Block. (The exempted services, which are within the Secretary of States responsibility, but not in the Scottish Block, are services where decisions are still essentially made at GB level- principally agriculture.) Table 1 shows the composition of that element of the control total relating to Scottish departments, and how this breaks down to give the Scottish block. 4.) Each year, as part of the annual Public Expenditure round, the Government determines any necessary changes to the overall control total, and to the public expenditure programmes of the individual Whng public expenditure control system: but the consequences will apply unchanged for the proposed funding arrangements for the new Scottish Parliament. (b) The squeeze on the Scottish Block implied by Barnett, and, in particular, the multiplier effect on council tax levels, will be extremely severe: and, indeed, likely to prove the dominant issue for the new Scottish Parliament. It is ironic that public debate on taxation for the new Scottish Parliament has focused on the (limited) ability of the Scottish Parliament to increase income tax. In practice, the public issue will be the continuing pressure on council tax levels. (c) The relative pressure on council tax levels in Scotland could be even more severe than implied by the above, if, (as is likely), the council tax base is more buoyant in England than in Scotland. Comment. 13.) The above analysis is purely factual, and is independent of whatever view is taken on what relative level of spending is needed to provide a similar overall level of service in Scotland and England. The whole question of spending needs is both complex and contentious. The only points I would like to make here are:- (a) The view that there is no extra spending need in Scotland relative to England is unlikely to be sustained in any objective study. For example, a Treasury study at the time of the last devolution Bill concluded that around 15% of extra spending per capita was needed in Scotland: (although the methodology used was arguable and suspect.) (b) Conversely, the view that all of Scotlands current relative level of spending can be justified on grounds of relative need is also unlikely to be sustainable. It is not difficult to point to service areas in Scotland where much more efficient provision ought to be achievable: and, indeed, it would be strange for the SNP to argue that the outcome of nearly three centuries of inefficient Westminster rule had nevertheless led to the most efficient provision of services possible. 14.) There is a widespread presumption that somehow the proposed continuation of Barnett represents a good deal for Scotland, which somehow preserves Scotlands favourable relative spending level: and the prospect of an exercise to establish relative spending needs between England and Scotland is, conversely, seen as a threat. On the basis of the analysis in this paper, this common view is perverse. The only rationale for the Barnett formula is the view that there is no extra relative spending need in Scotland: (and also implicit in Barnett, since it applies at the level of net expenditure, is the view that Scottish local authorities can raise the same level of per capita fees and charges as English- or otherwise Scottish gross spending levels will be forced to be correspondingly less.) Barnett will in due course force convergence in relative spending levels: and in the process, Scottish council tax levels will be put under extreme pressure. Faced with this prospect, the possibility of an exercise to attempt to establish objectively relative spending needs between Scotland and England should not necessarily be dismissed out of hand. The outcome of such an exercise would probably be that some level of extra spending in Scotland was justified on grounds of need, though probably below current levels. A planned transition to such an intermediate level might be a more manageable prospect for the Scottish Parliament than trying to cope with the greater eventual squeeze to be imposed by Barnett. 15.)There is a major policy issue for the SNP in the whole question of the Barnett squeeze. One point of view, for example, would be that all that the SNP has to do is to sit back, wait for the Barnett bomb to detonate itself under the Scottish Parliament, and then independence can be picked up out of the wreckage. I would argue that this point of view is fundamentally misplaced. As the Scottish parliament runs into increasing difficulty with programme cuts and council tax rises, the public perception is likely to be that made in Scotland policies are too expensive, and that a Sitehall spending departments. Typically, this involves a series of bilateral discussions between the relevant Minister, and the Chief Secretary to the Treasury. Changes to the Scottish Block, however, are determined simply by the application of an agreed formula: this is the so-called Barnett formula, introduced in 1978 to provide the basis for funding the devolution proposals then current. Under the Barnett formula, the Scottish Block is changed by the pro-rata population share of the change to the relevant English programme: or, putting this another way, the per capita increase to the Scottish Block is the same as the per capita increase to the English programme. 5.) The implication of the Barnett formula is that, if public expenditure levels are rising in money terms, then this will in due course cause convergence in the average levels of public expenditure per head in England and Scotland. (This can be readily seen, since, if the per capita expenditure on a given programme is higher in Scotland than EnglDDDDDDDDDDDE-E[ExEEEEEEF-FJFvFFFFGGGGJGxGyGGGG H9H^HHHHHH~Ip p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p -~IIIIIIIIIIIIIIIII6URUTUZZZZ]]]]]O`p p p p p p p p p p p p p `%K@Normala "A@"Default Paragraph Font @ Footer 9r )@ Page Number expenditure on services, and other items like general government debt interest. Twhilethe remainder is entirely for which he is responsibleT 2 .A. 4 4s in the UK. The control total includes the public expenditure programmes of the Governments service departments, e.g., Health, Education, etc.qw\#cottish legislature is incapable of controlling public expenditure. If this is the perception created by the financial difficulties of the Scottish Parliament, then it is not a basis from which the Scottish people will want to move on to independence. What is required is for the SNP to convince the Scottish public that (a) that the SNP is the party of financial responsibility, which can offer new and more efficient ways of delivering services: and also (b)that independence offers a more rational, and affordable funding mechanism for the Scottish Parliament. If the SNP can achieve both of these aims, then it should be able to use the financial problems of the Scottish Parliament as a springboard for independence: otherwise, if the SNP fails on either of these aims, then the case for independence will be lost. If this argument is accepted, then it poses an essential discipline on SNP policy in each service area. In each area, SNP policy must avoid simply being a wish list of all the desirable things money could buy: but instead, should stress, and demonstrate, how independence can offer new, better, and more efficient ways of delivering services. Table 1: The Scottish Block, 1995/96 Derivation of Block m Control Total, Scottish Departments 14300 less Forestry Commission 62 = Expenditure within SoS responsibility 14237 less Agriculture 521 less adjustment for nationalised industries -22 = Scottish Block 13738 Composition of Block Central Government Expenditure PIC  LMETA TCompObj fObjInfo Equation Native  Ole 1PIC  /LMETA %`7150 Central Govt support to local authority current expenditure 5483 Local Authority capital expenditure 984 Provision to match European fund receipts 120 = Scottish Block 13738 (Source: Serving Scotlands Needs: The Scottish Office, 1997) Table 2: Identifiable General Govt. Expenditure,1995/96: per head (excluding agriculture and social security) England Scotland Ratio, S/E Trade and Industry 79.6 130.9 1.64 Roads and Transport 173.3 163.9 0.95 Housing 68.3 14 FMicrosoft Equation 2.0 DS Equation Equation.29qV & .1   & s & MathType0 "- Times New Roman- 2 @y ` 2 = 2  ` 2 m[1 2 m-~2 m (.874)(.81~`~~` 2 m )]~ 2 .126` 2  x 2 k@ ```` 2 k,=2 kd2.32x` & "Systemn-LV(PmJ iJ4nJ y = PC E PC S . x FMicrosoft Equation 2.0 DS Equation Equation.29q  + .1  @& & MathType "-Times New Roman- 2 @y ` 2 = 2  ` 2 `PC 2 PC 2 r x`Times New Roman- 2 E 2 STimes New Roman- 2 .` & "Systemn-L 2.4 2.09 Other Envmental Services 150.6 263.8 1.75 Law, Order etc 247.3 267.3 1.08 Education 590.7 791.4 1.34 National Heritage 48.5 55.3 1.14 Health + Personal Soc. Serv 809.2 990.2 1.22 Total 2167.4 2802.6 1.29 (source: derived from Public Expenditure Statistical Analysis, 1997, table 7.7, using population figures from Regional trends 31) PAGE  PAGE 7  .A a quantity known as : this includes both, and other items such as general government debt interest. Note that the debt interest paid by general government, which includes the debt interest of local authorities, is not allocated to service departments, but is treated as a separate component of the control total., which comprises about 96% of the Scottish departments control total. ss aggregate of thes individual s, even if just on servicesSuppose that,programme: in the Scottish programme, which could be funded through the Barnett formula, arisingIIIIIIIIITTTTTTTUU4U5U6U7UPUQURUSUTUUUVU]U`UaUbUUUUUUUUUUUUUUXLX[XXXXXYGYHYXYYYZYaYkYlYxYyYYYYYYYYYYZZZZZvZZZZZZZZZZZuD8ceKuD8avK uDcPaucX, and other items such as general government debt interest. Note that the debt interest paid by general government, which includes the debt interest of local authorities, is not allocated to service departments, but is treated as a separate component of the control total., which comprises about 96% of the Scottish departments control total. ss aggregate of thes individual s, even if just on servicesSuppose that,programme: in the Scottish programme, which could be funded through the Barnett formula, arising fromdto the Scottish Block Thus, for example, for a major service like education, an x% increase in the programme in England will provide, via Barnett, sufficient funding for an increase of  EMBED Equation.2  in Scotland. This magnitude of squeeze on a major and important programme poses very difficult questions about priorities elsewhere in the Block. complete,anoting the following three local authorityfirst, which they may levy for the use of local authority facilities and certain services.his incomef for local authority current expenditure, (that is, by more than they are already being squeezed by Barnett). Since the major non-local authority area of spending is health, the scope for compensating reductions elsewhere in the Block will, in practice, be severely limited.tends to rise proportionately faster a analysis of the consequences of the Barnett formula: it will apply irrespective P the SNP, and how, April 1998. 7 .A 7BlockBlockBlock 1ZZ[[#[$[8[e[f[j[k[l[m[v[w[[[[[[[[\\.\\\\]<]>]?]X]c]q]s]]]]]]]]]]]]]]]]]`#`0`1`=`>`J`K`N`O`P`UcuPac>&)RgtPEFMNAKJ&D&..(.).*.+.>3F3S6[6DDDDE E$E,E[E_ExE|EEEEEFFGGHH}I~IIIIIIIIIIIIIIIII¼uPaP uDPuDy8ceKuDy8avK^cuDX8ceKuDX8avK uDcchcU^cUcI%RSTfguOP 5 6 v!yzM+IJ b"c"##$%p p p p p p p p p p p p p  p p p  p p  p p p p p p p p p p p p p p p p p p p p p p p p p +% & &E&F&e&h&&&&&}'~''((**--..-...j/k//0v2=3>3G3H34577==@AAA4B4CDDp p p p p p p p p p p p p p p p p  p p p p p p p p p p p p p p p p p p p p p  p p p p p p p -: for this reason, the figures are described as identifiable expenditure.: 4J]        %R/E={EJEz<%uQ|G /}lu!""|%&*-23>ACfEgEInJJp p  p p  p  p p p p p p p p p p p p p p p  p p p p  p p p p fromdto the Scottish Block Thus, for example, for a major service like education, an x% increase in the programme in England will provide, via Barnett, sufficient funding for an increase of  EMBED Equation.2  in Scotland. This magnitude of squeeze on a major and important programme poses very difficult questions about priorities elsewhere in the Block. complete,anoting the following three local authorityfirst, which they may levy for the use of local authority facilities and certain services.his incomef for local authority current expenditure, (that is, by more than they are already being squeezed by Barnett). Since the major non-local authority area of spending is health, the scope for compensating reductions elsewhere in the Block will, in practice, be severely limited.tends to rise proportionately faster a analysis of the consequences of the Barnett formula: it will apply irrespective P the SNP, and how, April 1998. 7 .A 7BlockBlockBlock 1ZZ[[#[$[8[e[f[j[k[l[m[v[w[[[[[[[[\\.\\\\]<]>]?]X]c]q]s]]]]]]]]]]]]]]]]]uPac3=J]        1%/=.E=JEI%uQ|G /3"r+!!e")S-235>|ACEEI!J=Jp p  p p  p  p p p p p p p p p p p p p  p p p p  p p p p p p  IZ]%+/%D~I]&'()I]_9-M-O-=J::: !!TV]e?D^`jmnq;A%%&&^+c+N-P-S-W-X-Z-....3388+>2>>>>>EE'F3FFFHHHHWI[I%JAPreferred Customer+C:\My Documents\Implications of Barnett.doc@HP DeskJet 870C SeriesLPT1:HPRDJC03HP DeskJet 870C SeriesHP DeskJet 870C Series7 d,,HP DeskJet 870C SeriesLPT1 ,,HP DeskJet 870C Series7 d,,HP DeskJet 870C SeriesLPT1 ,,8&456KLWY[j>[z*?D   x  " 0 (5BKbw!"y&';himnop!!~!!!!!!!" ###@#C#J#W#X#^+c+R,S,--.../e/r//F3j33333333334??ADDDDEE%E J J"J#J.J6J7J9J:J;J,APU@.-A\A.A]AA/A.\A\A\A\@i/A]@3A`3A<]Ab3A>]A?]AX]Ac]Aq]As]@3A]@E?@@AA]A6DA]ACD@]A]@DA4U@qI@]@IAIA]@I@I@I@I@I1Times New Roman Symbol &Arial"1hY$F\$f$fn !=!\$Implications of the Barnett Formula:Preferred CustomerPreferred Customerp p  p  p  p p p p p p p p p p p p p  p p p p  p p p p p p  IZ]%+/%D~I]&'()I]_p p  IZP`%+/%D~IO`&'()I]_#%---J::: !!TV]e;A%%&&>/E/'4.489B9x>>>>D?K?*F-FtFFFFHI,I/IIIpJJAPreferred Customer+C:\My Documents\Implications of Barnett.doc@HP DeskJet 870C SeriesLPT1:HPRDJC03HP DeskJet 870C SeriesHP DeskJet 870C Series7 d,,HP DeskJet 870C SeriesLPT1 ,,HP DeskJet 870C Series7 d,,HP DeskJet 870C SeriesLPT1 ,,%%%%&456KLWY[j>[z*?D   x  " 0 (5BKbw/R_`lmyzkl"#$%&45acpq `!a!!!!""%"*"#S#[#\######%%++,, .5.../\////33333444'45474S4@@ BDEEEeEgErEXJYJoJpJ{JJJJJJJJ@AUU@AVUA]UAAIAIAIAIAA`UAaUAbUAUAUAUAUAUAUAUAUAUAUAUAUAXALXA[XAXAXAXAX@ATAzAUAAUAAYAGYAHY@ A]@3 A A AXYA AYYAAZYAaYAsAkYAAlYAAxY@AyYAAYArAYAAYAAYA\AYAYAYAYAY@AZ@AZ@@Z@A`A#`A0`A1`A=`A>`AJ`@@ZAZAvZAZAZAZAZAZAZAZAZAZAZAZA[A[A#[A$[A8[Ae[Af[Aj[Ak[Al[@ A Am[A Av[@O"Aw[A"A[A"@"A[@#A[A[A[A#A[A[A&$A\@3$AK`@D&A]A>,APU@.-A\A.A]AA/A.\A\A\A\@i/A]@3A`3A<]Ab3A>]A?]AX]Ac]Aq]As]@3A]@E?@@AA]A6DA]ACD@]A]@DA4U@qI@N`@IAIAO`@I@I@I@I@I1Times New Roman Symbol &Arial"1hY$F8$f$fp a=!\$Implications of the Barnett Formula:Preferred CustomerPreferred Customer