PES are used as incentives to preserve the environment. Generally, the polluters or the users are the payers.

Ecosystem services are a public good and often partially or jointly owned by private land owners or holders of water rights, grazing permits, or forestry leases. Where afforestation, or reforestation, or wetland restoration, or beach re-nourishment are implemented public money is often used and private concerns are directly rewarded in either increased property values or continued profitable access to the permitted uses from which they gain some measure of commercial return.

So there is no simple answer of asking a utility to restore forests or wetlands at their own expense. Charging polluters for the remediation of polluted water in the Florida everglades has proved difficult due to litigation. Regulation of construction in the coastal high hazard zone has been impeded by the United States Supreme Court's decisions that ruled such regulation amounts to a violation of the "takings clause" of the Fifth Amendment and Fourteenth Amendments. While polluters ought to pay and thus a fund created out of which to reimburse the costs imposed on victims of pollution, I would like to see a case wherein this approach is working today.

Ecosystem services, when monetized, can be allocated among public and private beneficiaries through either insurance pools, land trusts, or covenants that protect rights in condominium. But the more diffuse the benefits to an ever wider public, the greater the difficulty in allocating "shares," to these beneficiaries. 

You are correct, the many market failures generated by pollution, land-use, and land-use changes must be accounted for in terms of the loss of ecosystem services. Yes, polluter should pay the victims as one would in a case of liability. Severance taxes could be one means of paying for the destruction of a water resource due to sand, rock, or surface coal mining. Another method proposed in Europe is carbon taxes on those who turn solid carbon into thermal-trapping carbon dioxide emissions or air acidifying sulfur pollution. In both cases the ecosystem services are often not counted, or subsumed under other private property-related assets. 

The problem is not unresolvable, but allocating ecosystem services, in a monetary and commercial world of exchange where carbon has not, as yet, been monetized will remain a challenging, inaccurate and less than fair process of determining actual losses and tangible assets.

Una manera de valorar el esfuerzo de recuperar áreas degradadas o esfuerzo de conservar estaría indicado por la Disposición a pagar DAP o WTP por lo tanto si el esquema de Pago por servicios ambientales recoge esa medida de DAP sería válido el análisis de que con el PES (correctamente valorado) representa ese esfuerzo. 

In Ecuador a program named Sociobosque (a PES scheme) pays poor peasants to conserve their land forest. Those PES made by governement with public money to private or common pool owners are made without a relevant economic measure, because its clear that market in this particular context wont determine prices of environmental goods and services or economic efforts of reclamation. People payed must received the "bono de la probreza" (poverty bonus). To know details visit this web page (in spanish):  http://sociobosque.ambiente.gob.ec/ 

I'm not sure I understand the first question. If it's whether PES can be used to encourage reclamation/restoration, the answer is certainly yes. Although most PES programs focus on conservation, there are also many PES programs that support restoration. One recent example is the Reflorestar Program in the Brazilian state of Espirito Santo, which offers a 3-year payment for restoration coupled with a longer-term payment for conservation of the resulting restored habitat. The advantage of using PES for restoration rather than a traditional cost-share subsidy is that there is much greater control over outcomes, given the conditionality of payments. A recent impact evaluation I conducted of a PES program aimed at restoring degraded pastures in Colombia shows it to have been very effective at inducing land use change, and that these changes have then persisted several years after the program ended.

The answer to the second question depends on the nature of the ES that would be generated thanks to the restoration. Where there are clearly identifiable users (as is usually the case for water services), ideally these users would make the payments. For other services, there may be existing markets that could be tapped (eg if carbon is being sequestered, though carbon markets have many problems and limitations). Where the ES are public goods that benefit many people, arranging for user payment may not be feasible. In such cases, public financing would be required.

(As a comment to the second answer above, I would note that not all ES are public goods. Some are rival in consumption -- notably in the case of water services. Indeed, the cases where PES is most interesting are precisely those where there are clear users, because then PES can be set up as a transaction between two parties rather than as a government program.)