Compulsory Acquisition Of Land And Compensation In .

C o m p u l s o ry A c q u i s i t i o n o f L a n d a n d C o m p e n s at i o n i n I n f r a s t r u c t u r e P r o j e c t sprivate activities on the basis that they maycontribute to national economic growth—asopposed to the more specific and tangible publicbenefits usually associated with eminent domain.A topical illustration can be found in the caseof investment by foreign firms in large-scaleagricultural holdings in developing countries. Thisphenomenon has attracted a lot of internationalattention, promoted by some as key to unleashing the agricultural potential of regions such assub-Saharan Africa, while others characterize itas “land grabbing” bringing few benefits to hostcommunities and putting local livelihoods at risk.Whatever the merits of the various sides in thisdebate (and those merits are likely to be highlycontext specific and depend on the nature ofthe investment), it is noteworthy that a numberof countries have included commercial agriculture amongst the public purposes justifying theexercise of compulsory acquisition, and have usedthat power as a tool for assembling land for largescale investments.2 Some countries, by contrast,have tended to eschew this approach in favorof encouraging voluntary negotiations betweeninvestors and local communities.3China provides a particularly interesting caseof compulsory acquisition used for channellingland into ultimately non-public uses. Althoughthe law stipulates that the state can acquirecollectively-owned rural land only in the publicinterest, this term is not defined. Under law, ruralland can only be converted to urban uses if it hasfirst been acquired by the state—although there isnow an active market for urban land use rights inChina, the options for rural users to sell to urbanusers are extremely limited. As a consequence,the exercise of compulsory acquisition is widespread on the edges of China’s exploding cities aslocal governments struggle to find land for urbanexpansion. The public interest in such situationsis implicitly defined to include actions deemed tobe necessary to support rational urban growth. Aspart of China’s ongoing efforts to address a widerange of problems associated with compulsoryacquisition, however, ways of limiting the conceptof public interest are being considered.23See Tanzania, Land Acquisition Act, 1967, section 4(1)(g).Although not specifically reflected in national law, thisapproach has been adopted by the Government of Ghana inthe context of the Ghana Commercial Agriculture Project,recently approved for financing by the World Bank. SeeWorld Bank, Ghana Commercial Agriculture Project, ProjectAppraisal Document, at 100.Public Private PartnershipsThese types of issues are not new—but they areincreasingly coming to the fore in the contextof public-private partnerships, particularly incountries where jurisprudence on the subject hasnot been as expansive as that found in the UnitedStates. In such contexts, governments may findthat the exercise of compulsory acquisition attimes encounters public scepticism, particularlywhere the claimed public benefits are indirect orspeculative. Some laws or proposed laws currentlyunder consideration deal with this by provisionsintended to “raise the bar” when private actorsare involved. A draft bill under consideration inIndia4, for example, would limit the exercise ofgovernment acquisition on behalf of a private orPPP proponent to a relatively small fraction of theoverall land needed for the enterprise—the restwould need to be acquired through private negotiation leading to voluntary sales. In such cases,the eminent domain power can be seen as a toolto deploy where the majority of landowners havewillingly consented to a land transfer, and only afew holdouts remain.Legislative mechanisms to limit scope of‘public purpose”There are other tools that legislators have used totry to ensure that the public purpose limitationhas some “teeth”, whether the ultimate user of theland is a public, private or PPP entity. In Kenya,for example, the proponent of a compulsory acquisition is required to provide credible evidencethat the benefit to the public of the acquisitionwill outweigh the hardships to those affected. Anumber of countries try to restrict subsequenttransfers and land use changes of land taken for aspecific public purpose, in order to ensure that thepublic purpose justification was genuine, and notsimply a disguise for facilitating future commercialtransfers. Thus, some laws or proposals requiregovernments to offer the land back to the originalowners if it is not used for the purpose for whichit was acquired5, or allow the original owners toshare in the profits if acquired land is transferredto an unanticipated private use.6456India, Draft National Land Acquisition and Rehabilitationand Resettlement Bill, 2011.See for example, Cambodia’s Law on Expropriation,2010, Article 9, which gives the original owner priority torepurchase expropriated property that is not used for theintended purpose.India, supra. Note 4, section 70.www.worldbank.org/ppp3

4PPP InsightsIII. COMPENSATIONIssues surrounding compensation for losses suffered—who-gets-what when government acquiresa piece of land—are typically the most complexand controversial aspects of compulsory acquisition. A long-standing principle in many jurisdictions is that compensation should be guided bythe objectives of “equity” and “equivalence”—thatis, the adequacy of compensation should be measured against the goal of ensuring that people areneither impoverished nor enriched (Keith, 2008).A variation on this standard view arguesthat it may be appropriate in some cases, particularly where a taking is occurring in the contextof a development project or program, to aimbeyond equivalence to improving the positionof those affected wherever possible. This is theprinciple articulated in the World Bank Policy onInvoluntary Resettlement: “Displaced personsshould be assisted in their efforts to improve theirlivelihoods and standards of living or at leastrestore them, in real terms, to pre-displacementlevels or to levels prevailing prior to the beginningof project implementation, whichever is higher.”7In either case, applying the principles ofcompensation in practice has always been anextremely complex challenge. Appreciation ofthis complexity has deepened as fuller and morenuanced views of the rights that people hold overland have taken root in many parts of the world.Indeed, one key insight supported by comparative analysis is that legal approaches developed inthe context of Europe or North America—whereland rights are generally standardized and welldefined, land markets function, and land recordsare reliable—have proven to be ill-equipped fordealing with many developing country contextswhere such attributes are less common.Compensation issues can be convenientlygrouped according to two overlapping sets ofquestions: who should receive compensation forwhat kind of loss; and how should the quantumand type of compensation be determined?7World Bank Operational Policy 4.12 on InvoluntaryResettlement, paragraph 2(c).a. Identifying compensableinterests and those whohold themUnder the classical model, a government acquiresan entire land parcel from a private owner andcompensates the owner for the ownership interestin the parcel, along with other elements of compensation discussed below. Yet in many places,particularly in developing countries, this simplemodel encounters a wide range of exceptionsand complications, often poorly or incompletelyaddressed in national legislation.Private rights over state landIn many countries, full private ownership ofland does not exist, and instead people holdland under long-term leaseholds, certificates ofoccupancy, concessions or other arrangementswhile the state retains ownership of the land. Insome such cases, state ownership is nominal, andprivate rights over state land are held and transacted for all practical purposes as if the land wereprivately owned. In other cases, the nature of theretained state interest may be more significant. Inany event, in applying the principles of compulsory acquisition, it is important to focus on theprivate rights to land that will be terminated asa result of a taking, whether or not such rightsamount to a narrow definition of ownership.Multiple layers of rightsThere may be multiple layers of rights held byany number of rights holders. A privately ownedparcel may be subject to leaseholds, mortgages,rights of way for utilities or transportation, concessions, rights of traditional or other uses, rightsto forest products, etc. Ownership of land, trees,buildings and other improvements may all beseparately held. Each of these separate interestsmay represent a significant loss to its holder if theland parcel is acquired by government and theright is terminated. Existing compulsory acquisition laws and practices may not be well adaptedto “catching” all relevant interests in a land parcel. Some laws may target land owners withoutmentioning the array of other potential rights thatmay also be relevant and affected by the acquisition. A more frequent scenario is that even wherethe legal framework clearly recognizes subsidiarywww.worldbank.org/ppp

C o m p u l s o ry A c q u i s i t i o n o f L a n d a n d C o m p e n s at i o n i n I n f r a s t r u c t u r e P r o j e c t sor secondary rights, the processes put in place foridentifying, notifying and compensating interestholders are not well-designed to discover the existence of such rights in a particular context, or tobring the holders of those rights into the discussion of compensation. This is particular the casewith respect to customary rights, discussed below.Unregistered or inaccuratelydocumented rightsIt is not unusual for compulsory acquisition lawsto presume a level of documentation of rightsthat may in fact not exist—once again perhaps alegacy of legal approaches from developed landmarket economies not being sufficiently adaptedto the realities of less developed contexts. Somelaws, for example, tie eligibility for compensationnarrowly to whether the land right is registeredin accordance with the country’s land registrationlegislation. This can be problematic where, asis frequently the case, only a fraction of a givencountry’s land has actually been registered. Manycountries have modern registration laws on thebooks, but implementation frequently suffers fromfinancial or other capacity constraints or a lackof political will. And in many cases registrationsystems may not capture all important secondaryrights that are present. In such contexts, too strictapplication of a “registered-interests-only” ruleto compensation would result in many interestsgoing uncompensated or under-compensated.To deal with this problem realistically canplace additional burdens on government. InAlbania, for example, it is reported that theDirectorate of Roads devotes a significant amountof time and resources to assisting unregisteredlandowners to register their land, just so that theland can then immediately be expropriated andthe owners compensated in accordance with theexpropriation legislation. In practice, governmentsmay in fact rely on alternate forms of evidence inimplementation of compulsory acquisition laws,relaxing the rule to accommodate facts on theground. A more desirable approach would be toexplore the possibility of building flexibility intolegislative design where needed.The issue of registration touches upon thebroader problem of official records that are inaccurate or out-of-date, again a common featurein many countries. A frequent constraint to thefair and efficient implementation of compulsoryacquisition laws is that geographic data maywrongly define the size or location of the parcel,and that legal data lists the wrong person as theholder of the land right.Customary rightsThe issue of how customary land rights—presentin many countries in Africa, Latin America, Asiaand the Pacific and elsewhere—should be treatedin relation to compulsory acquisition shares someof the same features of issues outlined in thepreceding paragraphs. Land within a customarytenure regime may be held according to variousrights configurations, with the group collectivelyholding some rights alongside an array of individual or household rights to individual parcelsor resources.Recent decades have witnessed a growingwillingness on the part of a number of nationalgovernments to accord some formal recognitionto customary rights. A clear example of this isGhana, where up to 80% of all land is recognizedin accordance with the Constitution as owned bytraditional communities. In a range of other countries, including Mozambique, Tanzania, Uganda,Burkina Faso, Mali, Philippines, Cambodia, Peruand elsewhere in Latin America, land laws havebeen adopted that provide for the recognition ofcustomary tenure regimes to varying degrees.In short, the distinction between customary andstatutory tenure is in a number of places lesssignificant than it used to be.On the one hand, this trend is encouragingbecause it promises more robust and equitabletreatment under law for land rights that governments have historically tended to ignore or onlyto acknowledge weakly. On the other hand, inmany countries, while the principle of recognition of customary rights is proclaimed, thelegal framework for such recognition is vaguelyarticulated and moves towards operationalizing ithave been ambivalent. As a result, the situationon the ground is often one where rights remainunclear and vulnerable. Compounding this isthe fact that many compulsory acquisition lawson the books pre-date recent land law reformsand are ill-suited to deal with issues such as thewww.worldbank.org/ppp5

6PPP Insightsvaluation and compensation for customary rights(as discussed below).8Shared resourcesLand held by groups or under some sort of collective arrangement can pose special challenges tothe application of the principles of compensation,both in customary and some non-customary tenure contexts. An example of the former is Ghana,where traditional authorities (referred to in someparts of the country as “stools” or “skins”) hold“allodial” (full) ownership to the land on behalfof their communities. A non-customary lawexample is the ownership of rural agriculturalland in China, where law recognizes the collectiveas owner. In such instances, the law treats thetraditional authority or the collective entity as theholder of the compensable interest in land, andrelies on them—implicitly or explicitly, depending on the country—to ensure that land rightscompensation is appropriately shared withinthe group. Yet in both contexts, there have beensignificant and growing concerns that deficienciesin intra-group governance structures have led tocompensation not reaching the actually displacedindividual or household. This has receivedparticular attention in recent years in China,where central government authorities have issuedincreasingly strong policy statements designedto ensure that land compensation paid under theLand Administration Law actually reaches the“land losing farmer.”Compensation for informal occupationThe payment of compensation for rights that arenot legally recognized raises difficult policy questions. The policies of international organizations8See Government of Botswana, Land Policy Review (2003),quoted in R. Knight, Statutory Recognition of CustomaryLand Rights in Africa, FAO (2010). “The present policy,under which holders of property rights under customarylaw on tribal land are entitled to receive less compensationthan holders of common law lease rights on state landand tribal land is unjust. A unified and fair system ofland acquisition and compensation should be establishedthat is applicable to all land and all people with propertyinterests in land. The scope of compensation offeredand the rights of those to be compensated under the TribalLand Act should be extended to achieve parity with theprovisions of the Acquisition of Property Act There is nojustification for the continuance of two separate systems ofland acquisition and compensation—one for user rights intribal land and one for all other sorts of rights in tribal land,in state land and in freehold land. The operation of the dualsystem penalizes the poor and benefits the well off.”for the most part include “squatters” and otherinformal occupants or users as among those entitled to received resettlement assistance, but this isa frequent area in which international norms andnational law diverge. Many governments object tothe idea that even those who are clearly occupying land illegally are entitled to some level ofcompensation. This, it is argued, creates perverseincentives for people to ignore rules when theyoccupy land, and rewards illegal behaviour to thedetriment of the rule of law. These objections arecompounded when the illegal occupiers in question are not poor and vulnerable but are relativelywell-off and well-connected investors. On theother hand, it needs to be recognized that a resettlement approach that focuses only on those withformal legal rights to their occupation could havedetrimental development consequences. Informaloccupation of land in many settings is not a matter of choice but of necessity, induced by poverty,exacerbated by inaccessible land markets andpoorly functioning planning regimes, and in somecases condoned and encouraged by authorities.Hence, while a full legislative embrace of thenotion that squatters should be compensated isperhaps unlikely to occur in most countries, thereis a growing trend on the part of governments toadjust law and practice to deal with the individualand societal consequences associated with thedisplacement of informal occupants.b. Defining the appropriatequantum and form ofcompensationMarket value is the benchmark found in mostcompulsory acquisition laws when it comes tothe calculation of compensation for an acquiredasset. Laws and constitutions may also refer morebroadly to principles such as “just” or “fair” compensation. Yet a key consideration that emergeswhen one surveys the wide variety of economic,social and cultural settings in which takings occuris that there is no universally appropriate methodfor calculating loss. To design compensation packages that will genuinely approach the objective ofensuring that people are no worse off than theywere before the taking requires careful tailoring tolocal realities. Thus, even laws from highly developed market economies generally recognize thatcompensation needs to go beyond the value ofwww.worldbank.org/ppp

C o m p u l s o ry A c q u i s i t i o n o f L a n d a n d C o m p e n s at i o n i n I n f r a s t r u c t u r e P r o j e c t sland and other assets, and to varying degrees contemplate compensation for losses associated withdisturbance, costs related to moving and transition, in some cases harm to business, etc9. Yetin many developing country settings, assemblingthe right compensation package may encounteran even more complex array of variables, as thefollowing paragraphs will highlight.Assessing the fair market value oflost assetsA common legislative approach is to define market value as the amount a willing buyer wouldpay a willing seller on the open market wheresome choice exists. There are several reasons thiscalculation might be difficult to make in a givensetting (particularly when it comes to land valuesas opposed to other non-land assets). In someareas, formal land markets may be non-existentor extremely thin, especially in rural areas. Somelegal systems disallow the sale of particularcategories of land or place severe limits on suchsales. Underlying rights, as we have seen, maybe poorly defined in terms of content or duration. In some cases, active informal markets mayexist that, if studied, could in theory reveal whatland is selling for in the area, but governmentsare generally reluctant to acknowledge officiallythe existence of such markets. And even whereformal land markets may exist, the absence ofan established, independent valuation professionand the tendency of buyers and sellers to understate prices in order to minimize taxation canconspire to make the ascertainment of marketvalue very difficult.In the face of these constraints, laws mayrely on a variety of proxies for land value. InChina, for example, the value of agriculturalland is determined by applying a multiplier(which varies from place to place) to the averageproductivity of the land over a three year period.In Albania and Ghana values are assigned to different categories of land on the basis of schedulesand maps prepared periodically by government.India’s proposed land acquisition bill definesmarket value by reference to the amounts statedin recent deeds for comparable land, but includesa multiplier as a frank acknowledgement of the9See for example South Australia, Land Acquisition Act,1969.routine gap between stated values and actual values.10 Although admittedly easier to apply, noneof the approaches are ideal, and may in somecases result in significant divergence between the“legal“ and “actual“ market value of land.Replacement cost vs. fair market valueInternational norms such as those promoted by the World Bank, IFC and others refer to“replacement cost” as the appropriate benchmarkfor valuation of assets. This is also a term found ina minority of national laws11. Where land marketsare robust, replacement cost and fair market valueshould be roughly equivalent. But as noted above,fair market value may be defined at the nationallevel in a way that does not ensure that compensation will be adequate to acquire equivalentassets. In such contexts, a focus on “replacementcosts” at least in theory shifts attention usefullyto the calculation of what it would really take in agiven market to replace lost assets. When it comesto non-land assets such as housing and otherimprovements, a replacement cost approach alsoensures that the depreciation of lost assets are nottaken into account in the calculation of compensation, and that transaction costs associated with thepurchase of new assets are covered.Land-for-land vs. cash compensationThe complexities associated with assigningrealistic monetary values to lost assets anddisplaced rights is symptomatic of the difficultiesof applying a “standard” compulsory acquisitionlegal framework to social situations in whichland is valued differently than it is in functioning market economies. In developed countries,modern takings law is predicated on the assumption that land is a fungible commodity. While thisassumption does not always hold true even insuch countries, it encounters numerous exceptions in the developing world, particularly in ruralareas. Some land rights that are critical to rurallivelihoods, such as rights of pasture or access toforest resources, may simply not be susceptible tomonetization—their loss may only be genuinelyaddressed through the provision of alternatives.1011Supra note 4, section 20.See for example Tanzania, Vil

Compulsory acquisition is the power of govern-ment to acquire private rights in land for a public purpose, without the willing consent of its owner or occupant (Keith, 2008). This power is known by a variety of names depending on a country’s legal traditions, including eminent domain, expropr