There are two uncontroversial themes running through the cases relating to the assessment of damages for injury to earning capacity. One is that in general it is desirable for precise evidence to be called as to what the plaintiff would have been likely to earn but for the injury and what the plaintiff is likely to earn after it. The second is that the failure to call such evidence does not necessarily result in selection of only a nil or nominal figure as damages for impaired earning capacity.
 Thus, in relation to the first of these themes, in Paff v Speed (1961) 105 CLR 549 at 559 Fullagar J said that the “usual method of proving damages under [this] head is by calling evidence to show what the plaintiff could probably have earned during the rest of his life if he had not been injured and what, if anything, he is now capable of earning”. The same is true where the defendant is seeking to demonstrate that the diminution of earning capacity is only partial. Barwick CJ in Arthur Robinson (Grafton) Pty Ltd v Carter (1968) 122 CLR 649 at 657 said of a plaintiff who had been rendered almost a quadriplegic:
“He has lost his earning capacity, so far as the evidence goes, all earning capacity. In this connexion, however, too little attention it seems to me is paid to the possibilities which have and will yet open up for paraplegics and quadriplegics: but this ought to be the subject of the evidence and not of mere suggestion on the part of judge or advocate.”
 The first theme was stated with the qualification “in general”, because there is authority that in some circumstances over-elaborate evidence is unhelpful. In J K Kealley v Jones  1 NSWLR 723 at 734-735 Moffitt P said:
“Sometimes, particularly in a jury trial, evidence of some claimed prospective earnings but for injury may distract rather than aid quantification of loss. A jury may well make a better assessment of the loss of a young law student debarred by injury from becoming a barrister by evidence such as in Gilles’ case [(1975) 49 ALJR 349] uncluttered by evidence of earnings made by selected leading senior counsel, with or without the aid of statistics as to success and failure rates of students and barristers, where such evidence is led in supposed aid of quantifying the lost opportunity of becoming leading counsel in the foggy future.”
This approach may rest on a general recognition that “the task of assessing damages in personal injuries cases should be kept as simple as possible”: Jongen v CSR Ltd (1992) Aust Torts Reports 81-192 at 61,713 per Anderson J.
“in seeking to quantify his damages, a plaintiff could be well advised to offer [evidence of wage levels] in many cases; and likewise a defendant, in seeking to cut down the damage, might similarly be well advised to tender such evidence; neither, in the absence of such evidence, could complain, to the same effect, at any quantification arrived at. This, however, is far from asserting that in the absence of such evidence only nominal damages is appropriate. …[W]here a plaintiff has suffered a significantly disabling injury which obviously affects the range and nature of the work he can, therefore, perform, a tribunal of fact can, without specific evidence as to what other persons with that kind of disability can earn, make a judgment and assessment, on a percentage basis or otherwise, of the value of the lost capacity.”
Where the plaintiff calls incomplete evidence and there is only a low award for diminution of earning capacity, it is difficult for the plaintiff to complain: Minchin v Public Curator of Queensland  ALR 91 at 93; Giorginis v Kastrati (1998) 49 SASR 371 at 375. But it does not follow that a substantial award in a case where the evidence is incomplete cannot survive appellate attack by the defendant: Luntz, Assessment of Damages for Personal Injury and Death, 3rd ed, para[1.9.28]. This is so for several reasons.
 First, damages to compensate for that part of reduced economic capacity which will be reflected in the future are sometimes analysed as being one type of “general damages”. Like other types of “general damages”, as Fullagar J said in Paff v Speed (1961) 105 CLR 549 at 559, they are “of their very nature, incapable of mathematical calculation and (although the expression is apt to be misleading) commonly very much ‘at large’. They are also at large in the sense that a jury has, in serious cases, a wide discretion in assessing them.” In Russell v J Hargreaves & Sons Pty Ltd (1956) 30 ALJ 533 at 534 Taylor J said:
“Since … it is impossible to prescribe fixed or precise standards for the translation into terms of money of physical injuries and pain and suffering and varying degrees of physical incapacity, it is inevitable that individual opinions as to what amount may be said to constitute full compensation in any particular case will vary. Indeed within the bounds of reasonableness they may vary greatly.”
The field is an “uncertain” one: Breska v Lysaghts Works Pty Ltd (1956) 74 WN (NSW) 168 at 169 per Street CJ.
 Secondly, strictly the issue does not turn on a comparison between what money the plaintiff would have earned apart from the injury and what money the plaintiff will earn after the injury. The compensable loss is not a loss of income but the loss of capacity to earn income in a manner productive of financial loss: Graham v Baker (1961) 106 CLR 340 at 347. The income earned before the injury is relevant, but only as an evidentiary aid in assessing damages for the loss of capacity to earn income: Paff v Speed (1961) 105 CLR 549 at 566 per Windeyer J. Evaluation of the worth of a loss of capacity to earn – of a lost chance to earn – is of its nature a more imprecise inquiry than calculation of a lost income. It rests on the hypothesis – that the plaintiff will have undiminished capacity – which has been rendered false by events. It does not depend on calculating the income from a particular career which is no longer possible, but in calculating the damage to a capacity to carry on various careers. It is an exercise in estimation of possibilities, not proof of probabilities. Luntz, Assessment of Damages for Personal Injury and Death, 3rd ed, para[1.9.18], said:
“it is not necessary for the plaintiff to establish the future loss with the same degree of precision as the present and past loss … The court is really being asked to estimate as best it can the future effect of the injuries from which the plaintiff has been proved to be suffering as a result of the defendant’s wrongful act.”
In Malec v J C Hutton Pty Ltd (1990) 169 CLR 638 at 639 Brennan and Dawson JJ said: “the ascertainment of earning capacity involves an evaluation of possibilities, not establishing a fact as a matter of history.” They approved Lord Diplock’s statement in Mallett v McMonagle  AC 166 at 176: “in assessing damages which depend upon its view as to what will happen in the future or would have happened in the future if something had not happened in the past, the court must make an estimate as to what are the chances that a particular thing would or could have happened and reflect those chances, whether or not they are more or less than even, in the amount of damages …”. The majority (Deane, Gaudron and McHugh JJ) in Malec v J C Hutton Pty Ltd said at 643 that when the law takes account of future or hypothetical events in assessing damages, it can only do so in terms of the degree of probability of those events occurring in a range from just above the speculative to just below the certain. The inquiry – the process of estimation of possibilities – is thus an imprecise and indeterminate one to be carried out within very broad parameters. The trier of fact may have to form conclusions on “slender materials”: Callaghan v Wm C Lynch Pty Ltd  NSWR 871 at 877 per Evatt CJ, Herron and Sugerman JJ. That language, unlike the reference to permitting “guess work or speculation” elsewhere in that judgment, was not criticised in Ivkovic v Australian Iron & Steel Ltd  SR (NSW) 598 at 607 per Manning J. However, Menzies J said that sometimes the assessment of damages involves “guess work rather than estimation”: Jones v Schiffman (1971) 124 CLR 303 at 308; see also Linsell v Robson  1 NSWLR 249 at 259 per Mahoney JA; Chaplin v Hicks  2 KB 786 at 792 per Vaughan Williams LJ. Lord Diplock described the factors underlying the assessment of damages for diminished earning capacity as “matters of prophecy or judicial guesses” in Paul v Rendell (1981) 34 ALR 569 at 578. The English position was summarised thus by Lloyd LJ in Foster v Tyne and Wear County Council  1 All ER 567 at 570:
“when it comes to estimating loss of earning capacity, there is no such thing as a conventional approach; there is no rule of thumb which can be applied. It would be so much easier if there were. But there is not. In each case the trial judge has to do his best to assess the plaintiff’s handicap, as an existing disability, by reference to what may happen in the future. As has been said so often, that is necessarily a matter of speculation; it is necessarily a matter of weighing up risks and chances in all the circumstances of a particular case. The very fact that the approach must necessarily be so speculative means, of course, that the occasions on which this court will feel justified in interfering with a judge’s assessment will be few and far between, for there is no established range or standard against which to measure the judge’s award.”
 Thirdly, the mere fact that the quantum of damages is difficult to assess does not mean that the plaintiff is only entitled to a nominal sum. This principle applies as much to the assessment of damages for impaired earning capacity in injured plaintiffs as it does to pecuniary loss caused by negligent advice (Bowen v Blair  VLR 398) or to loss in the form of the diminished value of damaged property (Wheeler v Riverside Coal Transport Co Pty Ltd  Qd R 113), or equitable damages (Talbot v General Television Corp Pty Ltd  VR 224 at 250-1), or damages for breach of contract (Fink v Fink (1946) 74 CLR 127 at 143). In the last case, Dixon and McTiernan JJ put the following general proposition: “Where there has been an actual loss of some sort, the common law does not permit difficulties of estimating the loss in money to defeat the only remedy it provided for breach of contract, an award of damages.” This was followed in Sellars v Adelaide Petroleum NL (1994) 185 CLR 332 at 349 per Mason CJ, Dawson, Toohey and Gaudron JJ. The same is true in tort. In Naylor v Yorkshire Electricity Board  AC 529 at 548 Lord Devlin said: “in the law of damages … difficulty in calculation is not ordinarily taken as a ground either for reducing or for increasing the award”. The court will be more ready to shoulder the burden of acting without specific evidence where that evidence is difficult to call. In Biggin & Co Ltd v Permanite Ltd  1 KB 422 at 438, a sale of goods case involving a recovery in respect of damaged goods, Devlin J said: “It is only that where precise evidence is obtainable the Court naturally expects to have it. Where it is not the Court must do the best it can”. As McPherson J said in Nilon v Bezzina  2 Qd R 420 at 424: “The degree of precision with which damages are to be proved is proportionate to the proof reasonably available”. The courts on occasion cite in related contexts Bowen LJ’s related but stricter observation in Ratcliffe v Evans  2 QB 524 at 532-523, an injurious falsehood case:
“In all actions … on the case where the damage actually done is the gist of the action, the character of the acts themselves which produce the damage, and the circumstances under which these acts are done, must regulate the degree of certainty and particularity with which the damage done ought to be stated and proved. As much certainty and particularity must be insisted on … in … proof of damage, as is reasonable, having regard to the circumstances and the nature of the acts themselves by which the damage is done.”
 The application of these principles is illustrated in a category of cases where the injury to a plaintiff who had the chance of a relatively high income if an appropriate tertiary training had been undertaken has prevented that training. In this category in particular the courts have not declined to make substantial awards to compensate for impaired economic capacity measured by valuing the lost chance to earn high income even though there was no evidence of the possible ranges of income.
 In Ashford v Ashford (1970) 44 ALJR 195 at 196, a case where there was no evidence of particular income levels which would have been attained but for the injury and which could be attained after it, but where it was clear that there had been a diminution in earning capacity, Barwick CJ said for reasons including “the difficulty in determining what was the actual diminution of the appellant’s earning capacity, the margins within which a trial judge might properly exercise his discretion in arriving at a verdict in this case were necessarily fairly wide.” The High Court restored a verdict of the trial judge of $90,000 general damages, which evidently included $25,000 for impaired earning capacity.